docs(chart): add inline comments for createBucketsHook resources

fix(s3api): ListObjects with trailing-slash prefix returns wrong results

When ListObjectsV2 is called with a prefix ending in "/" (e.g., "foo/"),
normalizePrefixMarker strips the trailing slash and splits into
dir="parent" and prefix="foo". The filer then lists entries matching
prefix "foo", which returns both directory "foo" and "foo1000".

The prefixEndsOnDelimiter guard correctly identifies directory "foo" as
the target and recurses into it, but then resets the guard to false.
The loop continues and incorrectly recurses into "foo1000" as well,
causing the listing to return objects from unrelated directories.

Fix: after recursing into the exact directory targeted by the
trailing-slash prefix, return immediately from the listing loop.
There is no reason to process sibling entries since the original
prefix specifically targeted one directory.

.github/workflows/container_dev.yml

@@ -32,7 +32,7 @@ jobs:
             org.opencontainers.image.vendor=Chris Lu
       -
         name: Set up QEMU
-        uses: docker/setup-qemu-action@c7c53464625b32c7a7e944ae62b3e17d2b600130 # v1
+        uses: docker/setup-qemu-action@ce360397dd3f832beb865e1373c09c0e9f86d70a # v1
       -
         name: Create BuildKit config
         run: |
@@ -42,28 +42,28 @@ jobs:
           EOF
       -
         name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # v1
+        uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd # v1
         with:
           buildkitd-flags: "--debug"
           buildkitd-config: /tmp/buildkitd.toml
       -
         name: Login to Docker Hub
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # v1
+        uses: docker/login-action@b45d80f862d83dbcd57f89517bcf500b2ab88fb2 # v1
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
       -
         name: Login to GHCR
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # v1
+        uses: docker/login-action@b45d80f862d83dbcd57f89517bcf500b2ab88fb2 # v1
         with:
           registry: ghcr.io
           username: ${{ secrets.GHCR_USERNAME }}
           password: ${{ secrets.GHCR_TOKEN }}
       -
         name: Build
-        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v2
+        uses: docker/build-push-action@d08e5c354a6adb9ed34480a06d141179aa583294 # v2
         with:
           context: ./docker
           push: ${{ github.event_name != 'pull_request' }}

.github/workflows/container_foundationdb_version.yml

@@ -126,14 +126,14 @@ jobs:
           echo "seaweedfs_ref=$seaweed" >> "$GITHUB_OUTPUT"
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
+        uses: docker/setup-qemu-action@v4
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Login to Docker Hub
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
@@ -150,7 +150,7 @@ jobs:
           fi
 
       - name: Build and push image
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v7
         with:
           context: ./docker
           push: ${{ github.event_name != 'pull_request' }}

.github/workflows/container_latest.yml

@@ -98,7 +98,7 @@ jobs:
             org.opencontainers.image.vendor=Chris Lu
       - name: Set up QEMU
         if: matrix.platform != 'amd64'
-        uses: docker/setup-qemu-action@v3
+        uses: docker/setup-qemu-action@v4
       - name: Create BuildKit config
         run: |
           cat > /tmp/buildkitd.toml <<EOF
@@ -106,25 +106,25 @@ jobs:
             mirrors = ["https://mirror.gcr.io"]
           EOF
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
         with:
           buildkitd-flags: "--debug"
           buildkitd-config: /tmp/buildkitd.toml
       - name: Login to Docker Hub
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
       - name: Login to GHCR
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           registry: ghcr.io
           username: ${{ secrets.GHCR_USERNAME }}
           password: ${{ secrets.GHCR_TOKEN }}
       - name: Build ${{ matrix.platform }} ${{ matrix.variant }}
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v7
         env:
           DOCKER_BUILDKIT: 1
         with:
@@ -180,12 +180,12 @@ jobs:
             ghcr.io/chrislusf/seaweedfs
           tags: type=raw,value=${{ github.event_name == 'workflow_dispatch' && github.event.inputs.image_tag || 'latest' }},suffix=${{ steps.config.outputs.tag_suffix }}
       - name: Login to Docker Hub
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
       - name: Login to GHCR
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           registry: ghcr.io
           username: ${{ secrets.GHCR_USERNAME }}

.github/workflows/container_release_foundationdb.yml

@@ -35,14 +35,14 @@ jobs:
             org.opencontainers.image.vendor=Chris Lu
       -
         name: Set up QEMU
-        uses: docker/setup-qemu-action@v3
+        uses: docker/setup-qemu-action@v4
       -
         name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
       -
         name: Login to Docker Hub
         if: github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
@@ -57,7 +57,7 @@ jobs:
           fi
       -
         name: Build
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v7
         with:
           context: ./docker
           push: ${{ github.event_name != 'pull_request' }}

.github/workflows/container_release_unified.yml

@@ -60,16 +60,16 @@ jobs:
             build_args: TAGS=5BytesOffset
             tag_suffix: _large_disk
             
-          # Full tags - amd64 only
+          # Full tags - multi-arch
           - variant: full
-            platforms: linux/amd64
+            platforms: linux/amd64,linux/arm64
             dockerfile: ./docker/Dockerfile.go_build
             build_args: TAGS=elastic,gocdk,rclone,sqlite,tarantool,tikv,ydb
             tag_suffix: _full
-            
-          # Large disk + full tags - amd64 only
+
+          # Large disk + full tags - multi-arch
           - variant: large_disk_full
-            platforms: linux/amd64
+            platforms: linux/amd64,linux/arm64
             dockerfile: ./docker/Dockerfile.go_build
             build_args: TAGS=5BytesOffset,elastic,gocdk,rclone,sqlite,tarantool,tikv,ydb
             tag_suffix: _large_disk_full
@@ -117,7 +117,7 @@ jobs:
             
       - name: Set up QEMU
         if: (github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant) && contains(matrix.platforms, 'arm')
-        uses: docker/setup-qemu-action@v3
+        uses: docker/setup-qemu-action@v4
         
       - name: Create BuildKit config
         if: github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant
@@ -129,20 +129,20 @@ jobs:
           
       - name: Set up Docker Buildx
         if: github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
         with:
           buildkitd-config: /tmp/buildkitd.toml
           
       - name: Login to Docker Hub
         if: (github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant) && github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
           
       - name: Login to GHCR
         if: (github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant) && github.event_name != 'pull_request'
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           registry: ghcr.io
           username: ${{ secrets.GHCR_USERNAME }}
@@ -150,7 +150,7 @@ jobs:
           
       - name: Build and push ${{ matrix.variant }}
         if: github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant
-        uses: docker/build-push-action@v6
+        uses: docker/build-push-action@v7
         env:
           DOCKER_BUILDKIT: 1
         with:
@@ -198,14 +198,14 @@ jobs:
     steps:
       - name: Login to Docker Hub
         if: github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
           
       - name: Login to GHCR
         if: github.event_name != 'workflow_dispatch' || github.event.inputs.variant == 'all' || github.event.inputs.variant == matrix.variant
-        uses: docker/login-action@v3
+        uses: docker/login-action@v4
         with:
           registry: ghcr.io
           username: ${{ secrets.GHCR_USERNAME }}

.github/workflows/container_rocksdb_version.yml

@@ -82,19 +82,19 @@ jobs:
           echo "seaweedfs_ref=$seaweed" >> "$GITHUB_OUTPUT"
 
       - name: Set up QEMU
-        uses: docker/setup-qemu-action@c7c53464625b32c7a7e944ae62b3e17d2b600130 # v1
+        uses: docker/setup-qemu-action@ce360397dd3f832beb865e1373c09c0e9f86d70a # v1
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@e468171a9de216ec08956ac3ada2f0791b6bd435 # v1
+        uses: docker/setup-buildx-action@4d04d5d9486b7bd6fa91e7baf45bbb4f8b9deedd # v1
 
       - name: Login to Docker Hub
-        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # v1
+        uses: docker/login-action@b45d80f862d83dbcd57f89517bcf500b2ab88fb2 # v1
         with:
           username: ${{ secrets.DOCKER_USERNAME }}
           password: ${{ secrets.DOCKER_PASSWORD }}
 
       - name: Build and push image
-        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v2
+        uses: docker/build-push-action@d08e5c354a6adb9ed34480a06d141179aa583294 # v2
         with:
           context: ./docker
           push: true

.github/workflows/deploy_telemetry.yml

@@ -26,7 +26,7 @@ jobs:
       - name: Set up Go
         uses: actions/setup-go@v6
         with:
-          go-version: '1.24'
+          go-version-file: 'go.mod'
 
       - name: Build Telemetry Server
         if: github.event_name == 'workflow_dispatch' && inputs.deploy

.github/workflows/depsreview.yml

@@ -11,4 +11,4 @@ jobs:
       - name: 'Checkout Repository'
         uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8
       - name: 'Dependency Review'
-        uses: actions/dependency-review-action@05fe4576374b728f0c523d6a13d64c25081e0803
+        uses: actions/dependency-review-action@2031cfc080254a8a887f58cffee85186f0e49e48

.github/workflows/e2e.yml

@@ -23,17 +23,16 @@ jobs:
     runs-on: ubuntu-22.04
     timeout-minutes: 30
     steps:
-    - name: Set up Go 1.x
-      uses: actions/setup-go@a5f9b05d2d216f63e13859e0d847461041025775 # v2
-      with:
-        go-version: ^1.13
-      id: go
-
     - name: Check out code into the Go module directory
-      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v2
+      uses: actions/checkout@v6
+
+    - name: Set up Go
+      uses: actions/setup-go@v6
+      with:
+        go-version-file: 'go.mod'
 
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v3
+      uses: docker/setup-buildx-action@v4
       
     - name: Cache Docker layers
       uses: actions/cache@v5

.github/workflows/ec-integration-tests.yml

@@ -18,7 +18,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
       id: go
 
     - name: Check out code into the Go module directory

.github/workflows/ec-integration.yml

@@ -28,7 +28,7 @@ jobs:
     - name: Set up Go
       uses: actions/setup-go@v6
       with:
-        go-version: '1.24'
+        go-version-file: 'go.mod'
         
     - name: Build weed binary
       run: |

.github/workflows/fuse-integration.yml

@@ -22,7 +22,6 @@ permissions:
   contents: read
 
 env:
-  GO_VERSION: '1.24'
   TEST_TIMEOUT: '45m'
 
 jobs:
@@ -35,10 +34,10 @@ jobs:
     - name: Checkout code
       uses: actions/checkout@v6
       
-    - name: Set up Go ${{ env.GO_VERSION }}
+    - name: Set up Go
       uses: actions/setup-go@v6
       with:
-        go-version: ${{ env.GO_VERSION }}
+        go-version-file: 'go.mod'
         
     - name: Install FUSE and dependencies
       run: |

.github/workflows/go.yml

@@ -19,13 +19,12 @@ jobs:
     name: Go Vet
     runs-on: ubuntu-latest
     steps:
-    - name: Set up Go 1.x
-      uses: actions/setup-go@a5f9b05d2d216f63e13859e0d847461041025775 # v2
-      with:
-        go-version: ^1.13
-      id: go
     - name: Check out code into the Go module directory
-      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v2
+      uses: actions/checkout@v6
+    - name: Set up Go
+      uses: actions/setup-go@v6
+      with:
+        go-version-file: 'go.mod'
     - name: Get dependencies
       run: |
         cd weed; go get -v -t -d ./...
@@ -42,13 +41,12 @@ jobs:
     name: Build
     runs-on: ubuntu-latest
     steps:
-    - name: Set up Go 1.x
-      uses: actions/setup-go@a5f9b05d2d216f63e13859e0d847461041025775 # v2
-      with:
-        go-version: ^1.13
-      id: go
     - name: Check out code into the Go module directory
-      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v2
+      uses: actions/checkout@v6
+    - name: Set up Go
+      uses: actions/setup-go@v6
+      with:
+        go-version-file: 'go.mod'
     - name: Build
       run: cd weed; go build -tags "elastic gocdk sqlite ydb tarantool tikv rclone" -v .
 
@@ -56,12 +54,11 @@ jobs:
     name: Test
     runs-on: ubuntu-latest
     steps:
-    - name: Set up Go 1.x
-      uses: actions/setup-go@a5f9b05d2d216f63e13859e0d847461041025775 # v2
-      with:
-        go-version: ^1.13
-      id: go
     - name: Check out code into the Go module directory
-      uses: actions/checkout@8e8c483db84b4bee98b60c0593521ed34d9990e8 # v2
+      uses: actions/checkout@v6
+    - name: Set up Go
+      uses: actions/setup-go@v6
+      with:
+        go-version-file: 'go.mod'
     - name: Test
       run: cd weed; go test -tags "elastic gocdk sqlite ydb tarantool tikv rclone" -v ./...

.github/workflows/kafka-quicktest.yml

@@ -26,14 +26,14 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
         cache: true
         cache-dependency-path: |
           **/go.sum
       id: go
 
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v3
+      uses: docker/setup-buildx-action@v4
 
     - name: Install dependencies
       run: |

.github/workflows/kafka-tests.yml

@@ -37,7 +37,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
       id: go
 
     - name: Check out code
@@ -82,7 +82,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
       id: go
 
     - name: Check out code
@@ -132,7 +132,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
         cache: true
         cache-dependency-path: |
           **/go.sum
@@ -311,7 +311,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
         cache: true
         cache-dependency-path: |
           **/go.sum
@@ -473,7 +473,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
         cache: true
         cache-dependency-path: |
           **/go.sum
@@ -631,7 +631,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
         cache: true
         cache-dependency-path: |
           **/go.sum
@@ -789,7 +789,7 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
       id: go
 
     - name: Check out code

.github/workflows/metadata-subscribe-tests.yml

@@ -30,7 +30,6 @@ permissions:
   contents: read
 
 env:
-  GO_VERSION: '1.24'
   TEST_TIMEOUT: '10m'
 
 jobs:
@@ -43,10 +42,10 @@ jobs:
     - name: Checkout code
       uses: actions/checkout@v6
       
-    - name: Set up Go ${{ env.GO_VERSION }}
+    - name: Set up Go
       uses: actions/setup-go@v6
       with:
-        go-version: ${{ env.GO_VERSION }}
+        go-version-file: 'go.mod'
         
     - name: Build SeaweedFS
       run: |

.github/workflows/postgres-tests.yml

@@ -25,14 +25,14 @@ jobs:
     - name: Set up Go 1.x
       uses: actions/setup-go@v6
       with:
-        go-version: ^1.24
+        go-version: ^1.25
       id: go
 
     - name: Check out code
       uses: actions/checkout@v6
 
     - name: Set up Docker Buildx
-      uses: docker/setup-buildx-action@v3
+      uses: docker/setup-buildx-action@v4
 
     - name: Cache Docker layers
       uses: actions/cache@v5

.github/workflows/s3-iam-tests.yml

@@ -5,6 +5,8 @@ on:
     paths:
       - 'weed/iam/**'
       - 'weed/s3api/**'
+      - 'weed/credential/**'
+      - 'weed/pb/**'
       - 'test/s3/iam/**'
       - '.github/workflows/s3-iam-tests.yml'
   push:
@@ -12,6 +14,8 @@ on:
     paths:
       - 'weed/iam/**'
       - 'weed/s3api/**'
+      - 'weed/credential/**'
+      - 'weed/pb/**'
       - 'test/s3/iam/**'
       - '.github/workflows/s3-iam-tests.yml'
   
@@ -80,7 +84,7 @@ jobs:
     timeout-minutes: 25
     strategy:
       matrix:
-        test-type: ["basic", "advanced", "policy-enforcement"]
+        test-type: ["basic", "advanced", "policy-enforcement", "group"]
         
     steps:
       - name: Check out code
@@ -117,7 +121,7 @@ jobs:
             "basic")
               echo "Running basic IAM functionality tests..."
               make clean setup start-services wait-for-services
-              go test -v -timeout 15m -run "TestS3IAMAuthentication|TestS3IAMBasicWorkflow|TestS3IAMTokenValidation|TestIAM" ./...
+              go test -v -timeout 15m -run "TestS3IAMAuthentication|TestS3IAMBasicWorkflow|TestS3IAMTokenValidation|TestIAMUserManagement|TestIAMAccessKeyManagement|TestIAMPolicyManagement" ./...
               ;;
             "advanced")
               echo "Running advanced IAM feature tests..."
@@ -129,6 +133,11 @@ jobs:
               make clean setup start-services wait-for-services
               go test -v -timeout 15m -run "TestS3IAMPolicyEnforcement|TestS3IAMBucketPolicy|TestS3IAMContextual" ./...
               ;;
+            "group")
+              echo "Running IAM group management tests..."
+              make clean setup start-services wait-for-services
+              go test -v -timeout 15m -run "TestIAMGroup" ./...
+              ;;
             *)
               echo "Unknown test type: ${{ matrix.test-type }}"
               exit 1

.github/workflows/s3-parquet-tests.yml

@@ -41,7 +41,7 @@ jobs:
       - name: Set up Go
         uses: actions/setup-go@v6
         with:
-          go-version: ^1.24
+          go-version: ^1.25
           cache: true
       
       - name: Set up Python ${{ matrix.python-version }}
@@ -148,7 +148,7 @@ jobs:
       - name: Set up Go
         uses: actions/setup-go@v6
         with:
-          go-version: ^1.24
+          go-version: ^1.25
           cache: true
       
       - name: Run Go unit tests

.github/workflows/s3-proxy-signature-tests.yml

@@ -29,7 +29,7 @@ jobs:
         id: go
 
       - name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Build SeaweedFS binary for Linux
         run: |

.github/workflows/s3-spark-tests.yml

@@ -33,7 +33,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Install SeaweedFS
         run: |

.github/workflows/s3-tables-tests.yml

@@ -144,7 +144,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Pre-pull Trino image
         run: docker pull trinodb/trino:479
@@ -271,7 +271,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Pre-pull Spark image
         run: docker pull apache/spark:3.5.1
@@ -337,7 +337,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Pre-pull RisingWave image
         run: |
@@ -405,7 +405,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Pre-pull Python image
         run: docker pull python:3
@@ -471,7 +471,7 @@ jobs:
         id: go
 
       - name: Set up Docker
-        uses: docker/setup-buildx-action@v3
+        uses: docker/setup-buildx-action@v4
 
       - name: Pre-pull Python image
         run: docker pull python:3

.github/workflows/sftp-tests.yml

@@ -24,7 +24,6 @@ permissions:
   contents: read
 
 env:
-  GO_VERSION: '1.24'
   TEST_TIMEOUT: '15m'
 
 jobs:
@@ -37,10 +36,10 @@ jobs:
     - name: Checkout code
       uses: actions/checkout@v6
       
-    - name: Set up Go ${{ env.GO_VERSION }}
+    - name: Set up Go
       uses: actions/setup-go@v6
       with:
-        go-version: ${{ env.GO_VERSION }}
+        go-version-file: 'go.mod'
         
     - name: Install dependencies
       run: |

.github/workflows/spark-integration-tests.yml

@@ -43,7 +43,7 @@ jobs:
     - name: Set up Go
       uses: actions/setup-go@v6
       with:
-        go-version: '1.24'
+        go-version-file: 'go.mod'
 
     - name: Build SeaweedFS binary
       run: |

.github/workflows/test-s3-over-https-using-awscli.yml

@@ -24,7 +24,7 @@ jobs:
 
       - uses: actions/setup-go@v6
         with:
-          go-version: ^1.24
+          go-version: ^1.25
 
       - name: Build SeaweedFS
         run: |

.github/workflows/volume-server-integration-tests.yml

@@ -28,7 +28,6 @@ permissions:
   contents: read
 
 env:
-  GO_VERSION: '1.24'
   TEST_TIMEOUT: '30m'
 
 jobs:
@@ -46,10 +45,10 @@ jobs:
       - name: Checkout code
         uses: actions/checkout@v6
 
-      - name: Set up Go ${{ env.GO_VERSION }}
+      - name: Set up Go
         uses: actions/setup-go@v6
         with:
-          go-version: ${{ env.GO_VERSION }}
+          go-version-file: 'go.mod'
 
       - name: Build SeaweedFS binary
         run: |

docker/Dockerfile.foundationdb_large

@@ -1,4 +1,4 @@
-FROM golang:1.24 AS builder
+FROM golang:1.25 AS builder
 
 RUN apt-get update && \
     apt-get install -y build-essential wget ca-certificates && \

docker/Dockerfile.go_build

@@ -1,4 +1,4 @@
-FROM golang:1.24-alpine as builder
+FROM golang:1.25-alpine AS builder
 RUN apk add git g++ fuse
 RUN mkdir -p /go/src/github.com/seaweedfs/
 ARG BRANCH=${BRANCH:-master}

docker/Dockerfile.rocksdb_dev_env

@@ -1,4 +1,4 @@
-FROM golang:1.24 AS builder
+FROM golang:1.25 AS builder
 
 RUN apt-get update
 RUN apt-get install -y build-essential libsnappy-dev zlib1g-dev libbz2-dev libgflags-dev liblz4-dev libzstd-dev

docker/Dockerfile.rocksdb_large

@@ -1,4 +1,4 @@
-FROM golang:1.24 AS builder
+FROM golang:1.25 AS builder
 
 RUN apt-get update
 RUN apt-get install -y build-essential libsnappy-dev zlib1g-dev libbz2-dev libgflags-dev liblz4-dev libzstd-dev

docker/compose/master-cloud.toml

@@ -7,6 +7,7 @@
 
 [master.maintenance]
 # periodically run these scripts are the same as running them from 'weed shell'
+# Scripts are skipped while an admin server is connected.
 scripts = """
   lock
   ec.encode -fullPercent=95 -quietFor=1h

go.mod

@@ -1,6 +1,6 @@
 module github.com/seaweedfs/seaweedfs
 
-go 1.24.9
+go 1.25.0
 
 require (
 	cloud.google.com/go v0.123.0 // indirect
@@ -63,7 +63,7 @@ require (
 	github.com/prometheus/client_golang v1.23.2
 	github.com/prometheus/client_model v0.6.2 // indirect
 	github.com/prometheus/common v0.67.2 // indirect
-	github.com/prometheus/procfs v0.19.2
+	github.com/prometheus/procfs v0.20.1
 	github.com/rcrowley/go-metrics v0.0.0-20201227073835-cf1acfcdf475 // indirect
 	github.com/remyoudompheng/bigfft v0.0.0-20230129092748-24d4a6f8daec // indirect
 	github.com/seaweedfs/goexif v1.0.3
@@ -87,7 +87,7 @@ require (
 	github.com/xdg-go/stringprep v1.0.4 // indirect
 	github.com/youmark/pkcs8 v0.0.0-20240726163527-a2c0da244d78 // indirect
 	go.etcd.io/etcd/client/v3 v3.6.7
-	go.mongodb.org/mongo-driver v1.17.6
+	go.mongodb.org/mongo-driver v1.17.9
 	go.opencensus.io v0.24.0 // indirect
 	gocloud.dev v0.45.0
 	gocloud.dev/pubsub/natspubsub v0.45.0
@@ -97,7 +97,7 @@ require (
 	golang.org/x/image v0.36.0
 	golang.org/x/net v0.49.0
 	golang.org/x/oauth2 v0.35.0
-	golang.org/x/sys v0.41.0
+	golang.org/x/sys v0.42.0
 	golang.org/x/text v0.34.0 // indirect
 	golang.org/x/tools v0.41.0 // indirect
 	golang.org/x/xerrors v0.0.0-20240903120638-7835f813f4da // indirect
@@ -124,13 +124,13 @@ require (
 	github.com/apple/foundationdb/bindings/go v0.0.0-20250911184653-27f7192f47c3
 	github.com/arangodb/go-driver v1.6.9
 	github.com/armon/go-metrics v0.4.1
-	github.com/aws/aws-sdk-go-v2 v1.41.1
+	github.com/aws/aws-sdk-go-v2 v1.41.3
 	github.com/aws/aws-sdk-go-v2/config v1.32.7
 	github.com/aws/aws-sdk-go-v2/credentials v1.19.7
 	github.com/aws/aws-sdk-go-v2/service/s3 v1.95.0
 	github.com/cognusion/imaging v1.0.2
 	github.com/fluent/fluent-logger-golang v1.10.1
-	github.com/getsentry/sentry-go v0.42.0
+	github.com/getsentry/sentry-go v0.43.0
 	github.com/go-ldap/ldap/v3 v3.4.12
 	github.com/golang-jwt/jwt/v5 v5.3.1
 	github.com/google/flatbuffers/go v0.0.0-20230108230133-3b8644d32c50
@@ -341,7 +341,7 @@ require (
 	github.com/aws/aws-sdk-go-v2/service/sso v1.30.9 // indirect
 	github.com/aws/aws-sdk-go-v2/service/ssooidc v1.35.13 // indirect
 	github.com/aws/aws-sdk-go-v2/service/sts v1.41.6
-	github.com/aws/smithy-go v1.24.0
+	github.com/aws/smithy-go v1.24.2
 	github.com/boltdb/bolt v1.3.1 // indirect
 	github.com/bradenaw/juniper v0.15.3 // indirect
 	github.com/bradfitz/iter v0.0.0-20191230175014-e8f45d346db8 // indirect

go.sum

@@ -716,8 +716,8 @@ github.com/armon/go-metrics v0.4.1/go.mod h1:E6amYzXo6aW1tqzoZGT755KkbgrJsSdpwZ+
 github.com/atomicgo/cursor v0.0.1/go.mod h1:cBON2QmmrysudxNBFthvMtN32r3jxVRIvzkUiF/RuIk=
 github.com/aws/aws-sdk-go v1.55.8 h1:JRmEUbU52aJQZ2AjX4q4Wu7t4uZjOu71uyNmaWlUkJQ=
 github.com/aws/aws-sdk-go v1.55.8/go.mod h1:ZkViS9AqA6otK+JBBNH2++sx1sgxrPKcSzPPvQkUtXk=
-github.com/aws/aws-sdk-go-v2 v1.41.1 h1:ABlyEARCDLN034NhxlRUSZr4l71mh+T5KAeGh6cerhU=
-github.com/aws/aws-sdk-go-v2 v1.41.1/go.mod h1:MayyLB8y+buD9hZqkCW3kX1AKq07Y5pXxtgB+rRFhz0=
+github.com/aws/aws-sdk-go-v2 v1.41.3 h1:4kQ/fa22KjDt13QCy1+bYADvdgcxpfH18f0zP542kZA=
+github.com/aws/aws-sdk-go-v2 v1.41.3/go.mod h1:mwsPRE8ceUUpiTgF7QmQIJ7lgsKUPQOUl3o72QBrE1o=
 github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.4 h1:489krEF9xIGkOaaX3CE/Be2uWjiXrkCH6gUX+bZA/BU=
 github.com/aws/aws-sdk-go-v2/aws/protocol/eventstream v1.7.4/go.mod h1:IOAPF6oT9KCsceNTvvYMNHy0+kMF8akOjeDvPENWxp4=
 github.com/aws/aws-sdk-go-v2/config v1.32.7 h1:vxUyWGUwmkQ2g19n7JY/9YL8MfAIl7bTesIUykECXmY=
@@ -758,8 +758,8 @@ github.com/aws/aws-sdk-go-v2/service/ssooidc v1.35.13 h1:gd84Omyu9JLriJVCbGApcLz
 github.com/aws/aws-sdk-go-v2/service/ssooidc v1.35.13/go.mod h1:sTGThjphYE4Ohw8vJiRStAcu3rbjtXRsdNB0TvZ5wwo=
 github.com/aws/aws-sdk-go-v2/service/sts v1.41.6 h1:5fFjR/ToSOzB2OQ/XqWpZBmNvmP/pJ1jOWYlFDJTjRQ=
 github.com/aws/aws-sdk-go-v2/service/sts v1.41.6/go.mod h1:qgFDZQSD/Kys7nJnVqYlWKnh0SSdMjAi0uSwON4wgYQ=
-github.com/aws/smithy-go v1.24.0 h1:LpilSUItNPFr1eY85RYgTIg5eIEPtvFbskaFcmmIUnk=
-github.com/aws/smithy-go v1.24.0/go.mod h1:LEj2LM3rBRQJxPZTB4KuzZkaZYnZPnvgIhb4pu07mx0=
+github.com/aws/smithy-go v1.24.2 h1:FzA3bu/nt/vDvmnkg+R8Xl46gmzEDam6mZ1hzmwXFng=
+github.com/aws/smithy-go v1.24.2/go.mod h1:YE2RhdIuDbA5E5bTdciG9KrW3+TiEONeUWCqxX9i1Fc=
 github.com/bahlo/generic-list-go v0.2.0 h1:5sz/EEAK+ls5wF+NeqDpk5+iNdMDXrh3z3nPnH1Wvgk=
 github.com/bahlo/generic-list-go v0.2.0/go.mod h1:2KvAjgMlE5NNynlg/5iLrrCCZ2+5xWbdbCW3pNTGyYg=
 github.com/bazelbuild/rules_go v0.46.0 h1:CTefzjN/D3Cdn3rkrM6qMWuQj59OBcuOjyIp3m4hZ7s=
@@ -1053,8 +1053,8 @@ github.com/gabriel-vasile/mimetype v1.4.11 h1:AQvxbp830wPhHTqc1u7nzoLT+ZFxGY7emj
 github.com/gabriel-vasile/mimetype v1.4.11/go.mod h1:d+9Oxyo1wTzWdyVUPMmXFvp4F9tea18J8ufA774AB3s=
 github.com/geoffgarside/ber v1.2.0 h1:/loowoRcs/MWLYmGX9QtIAbA+V/FrnVLsMMPhwiRm64=
 github.com/geoffgarside/ber v1.2.0/go.mod h1:jVPKeCbj6MvQZhwLYsGwaGI52oUorHoHKNecGT85ZCc=
-github.com/getsentry/sentry-go v0.42.0 h1:eeFMACuZTbUQf90RE8dE4tXeSe4CZyfvR1MBL7RLEt8=
-github.com/getsentry/sentry-go v0.42.0/go.mod h1:eRXCoh3uvmjQLY6qu63BjUZnaBu5L5WhMV1RwYO8W5s=
+github.com/getsentry/sentry-go v0.43.0 h1:XbXLpFicpo8HmBDaInk7dum18G9KSLcjZiyUKS+hLW4=
+github.com/getsentry/sentry-go v0.43.0/go.mod h1:XDotiNZbgf5U8bPDUAfvcFmOnMQQceESxyKaObSssW0=
 github.com/ghodss/yaml v1.0.0/go.mod h1:4dBDuWmgqj2HViK6kFavaiC9ZROes6MMH2rRYeMEF04=
 github.com/gin-contrib/sse v1.1.0 h1:n0w2GMuUpWDVp7qSpvze6fAu9iRxJY4Hmj6AmBOU05w=
 github.com/gin-contrib/sse v1.1.0/go.mod h1:hxRZ5gVpWMT7Z0B0gSNYqqsSCNIJMjzvm6fqCz9vjwM=
@@ -1769,8 +1769,8 @@ github.com/prometheus/procfs v0.0.2/go.mod h1:TjEm7ze935MbeOT/UhFTIMYKhuLP4wbCsT
 github.com/prometheus/procfs v0.0.8/go.mod h1:7Qr8sr6344vo1JqZ6HhLceV9o3AJ1Ff+GxbHq6oeK9A=
 github.com/prometheus/procfs v0.1.3/go.mod h1:lV6e/gmhEcM9IjHGsFOCxxuZ+z1YqCvr4OA4YeYWdaU=
 github.com/prometheus/procfs v0.6.0/go.mod h1:cz+aTbrPOrUb4q7XlbU9ygM+/jj0fzG6c1xBZuNvfVA=
-github.com/prometheus/procfs v0.19.2 h1:zUMhqEW66Ex7OXIiDkll3tl9a1ZdilUOd/F6ZXw4Vws=
-github.com/prometheus/procfs v0.19.2/go.mod h1:M0aotyiemPhBCM0z5w87kL22CxfcH05ZpYlu+b4J7mw=
+github.com/prometheus/procfs v0.20.1 h1:XwbrGOIplXW/AU3YhIhLODXMJYyC1isLFfYCsTEycfc=
+github.com/prometheus/procfs v0.20.1/go.mod h1:o9EMBZGRyvDrSPH1RqdxhojkuXstoe4UlK79eF5TGGo=
 github.com/pterm/pterm v0.12.27/go.mod h1:PhQ89w4i95rhgE+xedAoqous6K9X+r6aSOI2eFF7DZI=
 github.com/pterm/pterm v0.12.29/go.mod h1:WI3qxgvoQFFGKGjGnJR849gU0TsEOvKn5Q8LlY1U7lg=
 github.com/pterm/pterm v0.12.30/go.mod h1:MOqLIyMOgmTDz9yorcYbcw+HsgoZo3BQfg2wtl3HEFE=
@@ -2113,8 +2113,8 @@ go.etcd.io/etcd/client/pkg/v3 v3.6.7 h1:vvzgyozz46q+TyeGBuFzVuI53/yd133CHceNb/Ah
 go.etcd.io/etcd/client/pkg/v3 v3.6.7/go.mod h1:2IVulJ3FZ/czIGl9T4lMF1uxzrhRahLqe+hSgy+Kh7Q=
 go.etcd.io/etcd/client/v3 v3.6.7 h1:9WqA5RpIBtdMxAy1ukXLAdtg2pAxNqW5NUoO2wQrE6U=
 go.etcd.io/etcd/client/v3 v3.6.7/go.mod h1:2XfROY56AXnUqGsvl+6k29wrwsSbEh1lAouQB1vHpeE=
-go.mongodb.org/mongo-driver v1.17.6 h1:87JUG1wZfWsr6rIz3ZmpH90rL5tea7O3IHuSwHUpsss=
-go.mongodb.org/mongo-driver v1.17.6/go.mod h1:Hy04i7O2kC4RS06ZrhPRqj/u4DTYkFDAAccj+rVKqgQ=
+go.mongodb.org/mongo-driver v1.17.9 h1:IexDdCuuNJ3BHrELgBlyaH9p60JXAvdzWR128q+U5tU=
+go.mongodb.org/mongo-driver v1.17.9/go.mod h1:LlOhpH5NUEfhxcAwG0UEkMqwYcc4JU18gtCdGudk/tQ=
 go.opencensus.io v0.21.0/go.mod h1:mSImk1erAIZhrmZN+AvHh14ztQfjbGwt4TtuofqLduU=
 go.opencensus.io v0.22.0/go.mod h1:+kGneAE2xo2IficOXnaByMWTGM9T73dGwxeWcUqIpI8=
 go.opencensus.io v0.22.2/go.mod h1:yxeiOL68Rb0Xd1ddK5vPZ/oVn4vY4Ynel7k9FzqtOIw=
@@ -2509,8 +2509,8 @@ golang.org/x/sys v0.13.0/go.mod h1:oPkhp1MJrh7nUepCBck5+mAzfO9JrbApNNgaTdGDITg=
 golang.org/x/sys v0.17.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/sys v0.20.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
 golang.org/x/sys v0.28.0/go.mod h1:/VUhepiaJMQUp4+oa/7Zr1D23ma6VTLIYjOOTFZPUcA=
-golang.org/x/sys v0.41.0 h1:Ivj+2Cp/ylzLiEU89QhWblYnOE9zerudt9Ftecq2C6k=
-golang.org/x/sys v0.41.0/go.mod h1:OgkHotnGiDImocRcuBABYBEXf8A9a87e/uXjp9XT3ks=
+golang.org/x/sys v0.42.0 h1:omrd2nAlyT5ESRdCLYdm3+fMfNFE/+Rf4bDIQImRJeo=
+golang.org/x/sys v0.42.0/go.mod h1:4GL1E5IUh+htKOUEOaiffhrAeqysfVGipDYzABqnCmw=
 golang.org/x/telemetry v0.0.0-20240228155512-f48c80bd79b2/go.mod h1:TeRTkGYfJXctD9OcfyVLyj2J3IxLnKwHJR8f4D8a3YE=
 golang.org/x/telemetry v0.0.0-20260109210033-bd525da824e2 h1:O1cMQHRfwNpDfDJerqRoE2oD+AFlyid87D40L/OkkJo=
 golang.org/x/telemetry v0.0.0-20260109210033-bd525da824e2/go.mod h1:b7fPSJ0pKZ3ccUh8gnTONJxhn3c/PS6tyzQvyqw4iA8=

k8s/charts/seaweedfs/Chart.yaml

@@ -1,6 +1,6 @@
 apiVersion: v1
 description: SeaweedFS
 name: seaweedfs
-appVersion: "4.13"
+appVersion: "4.17"
 # Dev note: Trigger a helm chart release by `git tag -a helm-<version>`
-version: 4.0.413
+version: 4.17.0

k8s/charts/seaweedfs/openshift-values.yaml

@@ -23,8 +23,15 @@
 #
 # Adjust storageClass and sizes to match your cluster's available StorageClasses.
 # On OpenShift you can discover them with: oc get storageclass
-
+global:
+  enableReplication: true
+  #  replication type is XYZ:
+  # X number of replica in other data centers
+  # Y number of replica in other racks in the same data center
+  # Z number of replica in other servers in the same rack
+  replicationPlacement: "000" # no data replica
 master:
+  replicas: 1
   data:
     type: "persistentVolumeClaim"
     size: "10Gi"
@@ -49,6 +56,7 @@ master:
       type: RuntimeDefault
 
 volume:
+  replicas: 1
   dataDirs:
     - name: data1
       type: "persistentVolumeClaim"
@@ -75,6 +83,7 @@ volume:
       type: RuntimeDefault
 
 filer:
+  replicas: 1
   data:
     type: "persistentVolumeClaim"
     size: "25Gi"

k8s/charts/seaweedfs/templates/all-in-one/all-in-one-deployment.yaml

@@ -243,8 +243,7 @@ spec:
               {{- if $httpsPort }}
               -s3.port.https={{ $httpsPort }} \
               {{- end }}
-              -s3.cert.file=/usr/local/share/ca-certificates/client/tls.crt \
-              -s3.key.file=/usr/local/share/ca-certificates/client/tls.key \
+              {{ include "seaweedfs.s3.tlsArgs" (dict "root" . "prefix" "s3.") | nindent 14 }}
               {{- end }}
               {{- if or .Values.allInOne.s3.enableAuth .Values.s3.enableAuth .Values.filer.s3.enableAuth }}
               -s3.config=/etc/sw/s3/seaweedfs_s3_config \
@@ -346,6 +345,9 @@ spec:
             - name: client-cert
               mountPath: /usr/local/share/ca-certificates/client/
               readOnly: true
+            {{- if .Values.allInOne.s3.enabled }}
+            {{- include "seaweedfs.s3.tlsVolumeMount" . | nindent 12 }}
+            {{- end }}
             {{- end }}
             {{ tpl .Values.allInOne.extraVolumeMounts . | nindent 12 }}
           ports:
@@ -473,6 +475,9 @@ spec:
         - name: client-cert
           secret:
             secretName: {{ include "seaweedfs.fullname" . }}-client-cert
+        {{- if .Values.allInOne.s3.enabled }}
+        {{- include "seaweedfs.s3.tlsVolume" . | nindent 8 }}
+        {{- end }}
         {{- end }}
         {{ tpl .Values.allInOne.extraVolumes . | nindent 8 }}
       {{- if .Values.allInOne.nodeSelector }}

k8s/charts/seaweedfs/templates/all-in-one/all-in-one-service.yml

@@ -17,6 +17,9 @@ metadata:
 spec:
   type: {{ .Values.allInOne.service.type | default "ClusterIP" }}
   internalTrafficPolicy: {{ .Values.allInOne.service.internalTrafficPolicy | default "Cluster" }}
+  {{- if and (semverCompare ">=1.31-0" .Capabilities.KubeVersion.GitVersion) .Values.allInOne.s3.trafficDistribution }}
+  trafficDistribution: {{ include "seaweedfs.trafficDistribution" (dict "value" .Values.allInOne.s3.trafficDistribution "Capabilities" .Capabilities) }}
+  {{- end }}
   ports:
   # Master ports
   - name: "swfs-master"

k8s/charts/seaweedfs/templates/filer/filer-statefulset.yaml

@@ -200,8 +200,7 @@ spec:
               {{- if .Values.filer.s3.httpsPort }}
               -s3.port.https={{ .Values.filer.s3.httpsPort }} \
               {{- end }}
-              -s3.cert.file=/usr/local/share/ca-certificates/client/tls.crt \
-              -s3.key.file=/usr/local/share/ca-certificates/client/tls.key \
+              {{ include "seaweedfs.s3.tlsArgs" (dict "root" . "prefix" "s3.") | nindent 14 }}
               {{- end }}
               {{- if .Values.filer.s3.enableAuth }}
               -s3.config=/etc/sw/seaweedfs_s3_config \
@@ -254,6 +253,9 @@ spec:
             - name: client-cert
               readOnly: true
               mountPath: /usr/local/share/ca-certificates/client
+            {{- if .Values.filer.s3.enabled }}
+            {{- include "seaweedfs.s3.tlsVolumeMount" . | nindent 12 }}
+            {{- end }}
             {{- end }}
             {{ tpl .Values.filer.extraVolumeMounts . | nindent 12 | trim }}
           ports:
@@ -384,6 +386,9 @@ spec:
         - name: client-cert
           secret:
             secretName: {{ include "seaweedfs.fullname" . }}-client-cert
+        {{- if .Values.filer.s3.enabled }}
+        {{- include "seaweedfs.s3.tlsVolume" . | nindent 8 }}
+        {{- end }}
         {{- end }}
         {{ tpl .Values.filer.extraVolumes . | indent 8 | trim }}
       {{- if .Values.filer.nodeSelector }}

k8s/charts/seaweedfs/templates/s3/s3-deployment.yaml

@@ -127,8 +127,7 @@ spec:
               {{- if .Values.s3.httpsPort }}
               -port.https={{ .Values.s3.httpsPort }} \
               {{- end }}
-              -cert.file=/usr/local/share/ca-certificates/client/tls.crt \
-              -key.file=/usr/local/share/ca-certificates/client/tls.key \
+              {{ include "seaweedfs.s3.tlsArgs" (dict "root" . "prefix" "") | nindent 14 }}
               {{- end }}
               {{- if .Values.s3.domainName }}
               -domainName={{ .Values.s3.domainName }} \
@@ -176,6 +175,7 @@ spec:
             - name: client-cert
               readOnly: true
               mountPath: /usr/local/share/ca-certificates/client/
+            {{- include "seaweedfs.s3.tlsVolumeMount" . | nindent 12 }}
             {{- end }}
             {{ tpl .Values.s3.extraVolumeMounts . | nindent 12 | trim }}
           ports:
@@ -267,6 +267,7 @@ spec:
         - name: client-cert
           secret:
             secretName: {{ include "seaweedfs.fullname" . }}-client-cert
+        {{- include "seaweedfs.s3.tlsVolume" . | nindent 8 }}
         {{- end }}
         {{ tpl .Values.s3.extraVolumes . | indent 8 | trim }}
       {{- if .Values.s3.nodeSelector }}

k8s/charts/seaweedfs/templates/s3/s3-service.yaml

@@ -16,8 +16,9 @@ metadata:
 {{- end }}
 spec:
   internalTrafficPolicy: {{ .Values.s3.internalTrafficPolicy | default "Cluster" }}
-  {{- if and (semverCompare ">=1.31-0" .Capabilities.KubeVersion.GitVersion) (or .Values.s3.trafficDistribution .Values.filer.s3.trafficDistribution) }}
-  trafficDistribution: {{ include "seaweedfs.trafficDistribution" . }}
+  {{- $td := .Values.s3.trafficDistribution | default .Values.filer.s3.trafficDistribution }}
+  {{- if and (semverCompare ">=1.31-0" .Capabilities.KubeVersion.GitVersion) $td }}
+  trafficDistribution: {{ include "seaweedfs.trafficDistribution" (dict "value" $td "Capabilities" .Capabilities) }}
   {{- end }}
   ports:
   - name: "swfs-s3"

k8s/charts/seaweedfs/templates/shared/_helpers.tpl

@@ -338,11 +338,41 @@ Create the name of the service account to use
 {{- .Values.global.serviceAccountName | default "seaweedfs" -}}
 {{- end -}}
 
-{{/* Generate a compatible trafficDistribution value due to "PreferClose" fast deprecation in k8s v1.35 */}}
+{{/* S3 TLS cert/key arguments, using custom secret if s3.tlsSecret is set */}}
+{{- define "seaweedfs.s3.tlsArgs" -}}
+{{- $prefix := .prefix -}}
+{{- $root := .root -}}
+{{- if $root.Values.s3.tlsSecret -}}
+-{{ $prefix }}cert.file=/usr/local/share/ca-certificates/s3/tls.crt \
+-{{ $prefix }}key.file=/usr/local/share/ca-certificates/s3/tls.key \
+{{- else -}}
+-{{ $prefix }}cert.file=/usr/local/share/ca-certificates/client/tls.crt \
+-{{ $prefix }}key.file=/usr/local/share/ca-certificates/client/tls.key \
+{{- end -}}
+{{- end -}}
+
+{{/* S3 custom TLS volume mount */}}
+{{- define "seaweedfs.s3.tlsVolumeMount" -}}
+{{- if .Values.s3.tlsSecret }}
+- name: s3-tls-cert
+  readOnly: true
+  mountPath: /usr/local/share/ca-certificates/s3/
+{{- end }}
+{{- end -}}
+
+{{/* S3 custom TLS volume */}}
+{{- define "seaweedfs.s3.tlsVolume" -}}
+{{- if .Values.s3.tlsSecret }}
+- name: s3-tls-cert
+  secret:
+    secretName: {{ .Values.s3.tlsSecret }}
+{{- end }}
+{{- end -}}
+
+{{/* Generate a compatible trafficDistribution value due to "PreferClose" fast deprecation in k8s v1.35.
+     Accepts a dict with "value" (the trafficDistribution string) and "Capabilities". */}}
 {{- define "seaweedfs.trafficDistribution" -}}
-{{- if .Values.s3.trafficDistribution -}}
-{{- and (eq .Values.s3.trafficDistribution "PreferClose") (semverCompare ">=1.35-0" .Capabilities.KubeVersion.GitVersion) | ternary "PreferSameZone" .Values.s3.trafficDistribution -}}
-{{- else if .Values.filer.s3.trafficDistribution -}}
-{{- and (eq .Values.filer.s3.trafficDistribution "PreferClose") (semverCompare ">=1.35-0" .Capabilities.KubeVersion.GitVersion) | ternary "PreferSameZone" .Values.filer.s3.trafficDistribution -}}
+{{- if .value -}}
+{{- and (eq .value "PreferClose") (semverCompare ">=1.35-0" .Capabilities.KubeVersion.GitVersion) | ternary "PreferSameZone" .value -}}
 {{- end -}}
 {{- end -}}

k8s/charts/seaweedfs/templates/shared/post-install-bucket-hook.yaml

@@ -51,7 +51,7 @@ metadata:
   annotations:
     "helm.sh/hook": post-install,post-upgrade
     "helm.sh/hook-weight": "-5"
-    "helm.sh/hook-delete-policy": hook-succeeded
+    "helm.sh/hook-delete-policy": before-hook-creation,hook-succeeded
 spec:
   template:
     metadata:
@@ -64,9 +64,11 @@ spec:
       {{- if .Values.filer.podSecurityContext.enabled }}
       securityContext: {{- omit .Values.filer.podSecurityContext "enabled" | toYaml | nindent 8 }}
       {{- end }}
+      {{- include "seaweedfs.imagePullSecrets" $ | nindent 6 }}
       containers:
       - name: post-install-job
         image: {{ template "master.image" . }}
+        imagePullPolicy: {{ $.Values.global.imagePullPolicy | default "IfNotPresent" }}
         env:
           - name: WEED_CLUSTER_DEFAULT
             value: "sw"
@@ -92,6 +94,7 @@ spec:
           - "/bin/sh"
           - "-ec"
           - |
+            set -o pipefail
             wait_for_service() {
               local url=$1
               local max_attempts=60  # 5 minutes total (5s * 60)
@@ -117,8 +120,7 @@ spec:
             wait_for_service "http://$WEED_CLUSTER_SW_MASTER{{ .Values.master.readinessProbe.httpGet.path }}"
             wait_for_service "http://$WEED_CLUSTER_SW_FILER{{ .Values.filer.readinessProbe.httpGet.path }}"
             {{- end }}
-          set -o pipefail
-          {{- range $createBuckets }}
+            {{- range $createBuckets }}
             {{- $bucketName := .name }}
             {{- $bucketLock := or .lock .objectLock .withLock }}
             bucket_list=$(/bin/echo 's3.bucket.list' | /usr/bin/weed shell) || { echo "Error listing s3 buckets"; exit 1; }
@@ -187,6 +189,10 @@ spec:
           {{- end }}
           - containerPort: {{ .Values.master.grpcPort }}
             #name: swfs-master-grpc
+        {{- with coalesce .Values.allInOne.s3.createBucketsHook.resources .Values.s3.createBucketsHook.resources .Values.filer.s3.createBucketsHook.resources }}
+        resources:
+          {{- toYaml . | nindent 10 }}
+        {{- end }}
         {{- if .Values.filer.containerSecurityContext.enabled }}
         securityContext: {{- omit .Values.filer.containerSecurityContext "enabled" | toYaml | nindent 12 }}
         {{- end }}

k8s/charts/seaweedfs/values.yaml

@@ -909,6 +909,10 @@ filer:
     #     versioning: Enabled
     #   - name: bucket-b
     #     anonymousRead: false
+    # Resource requests/limits for the post-install bucket creation hook job.
+    # Should map directly to the value of the resources field for a PodSpec.
+    createBucketsHook:
+      resources: {}
 
 s3:
   enabled: false
@@ -919,6 +923,13 @@ s3:
   port: 8333
   # add additional https port
   httpsPort: 0
+  # Use a custom TLS certificate secret for the S3 HTTPS endpoint.
+  # When set, this Kubernetes Secret (must contain tls.crt and tls.key) is used
+  # instead of the internal self-signed client certificate generated by cert-manager.
+  # This allows using a publicly trusted certificate (e.g., from Let's Encrypt)
+  # so that S3 clients don't need to trust the internal CA.
+  # Requires global.enableSecurity to be true.
+  tlsSecret: null
   metricsPort: 9327
   # Iceberg catalog REST port (Apache Iceberg REST Catalog API)
   # Set to a port number to enable, or 0/null to disable
@@ -1069,6 +1080,11 @@ s3:
     failureThreshold: 100
     timeoutSeconds: 10
 
+  # Resource requests/limits for the post-install bucket creation hook job.
+  # Should map directly to the value of the resources field for a PodSpec.
+  createBucketsHook:
+    resources: {}
+
   ingress:
     enabled: false
     className: ""
@@ -1453,6 +1469,7 @@ allInOne:
     # The s3-secret.yaml template only reads from .Values.s3.credentials.
     # See: s3.credentials.admin.accessKey, s3.credentials.read.accessKey
     auditLogConfig: null  # S3 audit log configuration (null inherits from s3.auditLogConfig)
+    trafficDistribution: null  # Service traffic distribution (e.g., "PreferClose"); auto-converts to "PreferSameZone" on k8s >=1.35
     # You may specify buckets to be created during the install process.
     # Buckets may be exposed publicly by setting `anonymousRead` to `true`
     # ttl format: [1-255][m|h|d|w|M|y] (e.g., 7d)
@@ -1466,6 +1483,10 @@ allInOne:
     #     versioning: Enabled
     #   - name: bucket-b
     #     anonymousRead: false
+    # Resource requests/limits for the post-install bucket creation hook job.
+    # Should map directly to the value of the resources field for a PodSpec.
+    createBucketsHook:
+      resources: {}
 
   # SFTP server configuration
   # Note: Most parameters below default to null, which means they inherit from

other/java/client/src/main/proto/filer.proto

@@ -100,10 +100,12 @@ message ListEntriesRequest {
     string startFromFileName = 3;
     bool inclusiveStartFrom = 4;
     uint32 limit = 5;
+    int64 snapshot_ts_ns = 6;
 }
 
 message ListEntriesResponse {
     Entry entry = 1;
+    int64 snapshot_ts_ns = 2;
 }
 
 message RemoteEntry {
@@ -203,6 +205,7 @@ message CreateEntryRequest {
 
 message CreateEntryResponse {
     string error = 1;
+    SubscribeMetadataResponse metadata_event = 2;
 }
 
 message UpdateEntryRequest {
@@ -212,6 +215,7 @@ message UpdateEntryRequest {
     repeated int32 signatures = 4;
 }
 message UpdateEntryResponse {
+    SubscribeMetadataResponse metadata_event = 1;
 }
 
 message AppendToEntryRequest {
@@ -236,6 +240,7 @@ message DeleteEntryRequest {
 
 message DeleteEntryResponse {
     string error = 1;
+    SubscribeMetadataResponse metadata_event = 2;
 }
 
 message AtomicRenameEntryRequest {
@@ -469,6 +474,7 @@ message CacheRemoteObjectToLocalClusterRequest {
 }
 message CacheRemoteObjectToLocalClusterResponse {
     Entry entry = 1;
+    SubscribeMetadataResponse metadata_event = 2;
 }
 
 /////////////////////////

seaweedfs-rdma-sidecar/Dockerfile.sidecar

@@ -1,5 +1,5 @@
 # Multi-stage build for Go Sidecar
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git ca-certificates tzdata

seaweedfs-rdma-sidecar/Dockerfile.test-client

@@ -1,5 +1,5 @@
 # Multi-stage build for Test Client
-FROM golang:1.23-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git ca-certificates tzdata

test/erasure_coding/admin_dockertest/ec_integration_test.go

@@ -153,6 +153,38 @@ func waitForUrl(t *testing.T, url string, retries int) {
 	t.Fatalf("Timeout waiting for %s", url)
 }
 
+func fetchJSON(url string, out interface{}) error {
+	resp, err := http.Get(url)
+	if err != nil {
+		return err
+	}
+	defer resp.Body.Close()
+	if resp.StatusCode != http.StatusOK {
+		body, _ := io.ReadAll(resp.Body)
+		return fmt.Errorf("status %d: %s", resp.StatusCode, string(body))
+	}
+	return json.NewDecoder(resp.Body).Decode(out)
+}
+
+func mapField(obj map[string]interface{}, key string) (interface{}, bool) {
+	if obj == nil {
+		return nil, false
+	}
+	if value, ok := obj[key]; ok {
+		return value, true
+	}
+	return nil, false
+}
+
+func mapFieldAny(obj map[string]interface{}, keys ...string) (interface{}, bool) {
+	for _, key := range keys {
+		if value, ok := mapField(obj, key); ok {
+			return value, true
+		}
+	}
+	return nil, false
+}
+
 func TestEcEndToEnd(t *testing.T) {
 	defer cleanup()
 	ensureEnvironment(t)
@@ -275,6 +307,7 @@ func TestEcEndToEnd(t *testing.T) {
 	startTime := time.Now()
 	ecVerified := false
 	var lastBody []byte
+	debugTick := 0
 
 	for time.Since(startTime) < 300*time.Second {
 		// 3.1 Check Master Topology
@@ -300,25 +333,104 @@ func TestEcEndToEnd(t *testing.T) {
 			}
 		}
 
-		// 3.2 Debug: Check workers and jobs
-		wResp, wErr := http.Get(AdminUrl + "/api/plugin/workers")
+		// 3.2 Debug: Check workers, jobs, and scheduler status
+		debugTick++
+
+		var workers []map[string]interface{}
 		workerCount := 0
-		if wErr == nil {
-			var workers []interface{}
-			json.NewDecoder(wResp.Body).Decode(&workers)
-			wResp.Body.Close()
+		ecDetectorCount := 0
+		ecExecutorCount := 0
+		if err := fetchJSON(AdminUrl+"/api/plugin/workers", &workers); err == nil {
 			workerCount = len(workers)
+			for _, worker := range workers {
+				capsValue, ok := mapFieldAny(worker, "capabilities", "Capabilities")
+				if !ok {
+					continue
+				}
+				caps, ok := capsValue.(map[string]interface{})
+				if !ok {
+					continue
+				}
+				if capValue, ok := caps["erasure_coding"].(map[string]interface{}); ok {
+					if capValue["can_detect"] == true {
+						ecDetectorCount++
+					}
+					if capValue["can_execute"] == true {
+						ecExecutorCount++
+					}
+				}
+			}
 		}
 
-		tResp, tErr := http.Get(AdminUrl + "/api/plugin/jobs?limit=1000")
+		var tasks []map[string]interface{}
 		taskCount := 0
-		if tErr == nil {
-			var tasks []interface{}
-			json.NewDecoder(tResp.Body).Decode(&tasks)
-			tResp.Body.Close()
+		ecTaskCount := 0
+		ecTaskStates := map[string]int{}
+		if err := fetchJSON(AdminUrl+"/api/plugin/jobs?limit=1000", &tasks); err == nil {
 			taskCount = len(tasks)
+			for _, task := range tasks {
+				jobType, _ := task["job_type"].(string)
+				state, _ := task["state"].(string)
+				if jobType == "erasure_coding" {
+					ecTaskCount++
+					ecTaskStates[state]++
+				}
+			}
+		}
+
+		t.Logf("Waiting for EC... (Workers: %d det=%d exec=%d, Tasks: %d ec=%d, EC States: %+v)",
+			workerCount, ecDetectorCount, ecExecutorCount, taskCount, ecTaskCount, ecTaskStates)
+
+		if debugTick%3 == 0 {
+			var pluginStatus map[string]interface{}
+			if err := fetchJSON(AdminUrl+"/api/plugin/status", &pluginStatus); err == nil {
+				t.Logf("Plugin status: enabled=%v worker_count=%v worker_grpc_port=%v configured=%v",
+					pluginStatus["enabled"], pluginStatus["worker_count"], pluginStatus["worker_grpc_port"], pluginStatus["configured"])
+			}
+
+			var schedulerStatus map[string]interface{}
+			if err := fetchJSON(AdminUrl+"/api/plugin/scheduler-status", &schedulerStatus); err == nil {
+				if schedValue, ok := schedulerStatus["scheduler"].(map[string]interface{}); ok {
+					t.Logf("Scheduler status: current_job_type=%v phase=%v last_iteration_had_jobs=%v idle_sleep_seconds=%v last_iteration_done_at=%v next_detection_at=%v",
+						schedValue["current_job_type"], schedValue["current_phase"],
+						schedValue["last_iteration_had_jobs"], schedValue["idle_sleep_seconds"], schedValue["last_iteration_done_at"], schedValue["next_detection_at"])
+				} else {
+					t.Logf("Scheduler status: %v", schedulerStatus)
+				}
+			}
+
+			var schedulerStates []map[string]interface{}
+			if err := fetchJSON(AdminUrl+"/api/plugin/scheduler-states", &schedulerStates); err == nil {
+				for _, state := range schedulerStates {
+					if state["job_type"] == "erasure_coding" {
+						t.Logf("EC scheduler state: enabled=%v detection_in_flight=%v detector_available=%v executor_workers=%v next_detection_at=%v last_run_status=%v last_run_started_at=%v last_run_completed_at=%v",
+							state["enabled"], state["detection_in_flight"], state["detector_available"],
+							state["executor_worker_count"], state["next_detection_at"], state["last_run_status"],
+							state["last_run_started_at"], state["last_run_completed_at"])
+						break
+					}
+				}
+			}
+
+			var jobTypes []map[string]interface{}
+			if err := fetchJSON(AdminUrl+"/api/plugin/job-types", &jobTypes); err == nil {
+				var names []string
+				for _, jobType := range jobTypes {
+					if name, ok := jobType["job_type"].(string); ok && name != "" {
+						names = append(names, name)
+					}
+				}
+				t.Logf("Plugin job types: %v", names)
+			}
+
+			var activities []map[string]interface{}
+			if err := fetchJSON(AdminUrl+"/api/plugin/activities?job_type=erasure_coding&limit=5", &activities); err == nil {
+				for i := len(activities) - 1; i >= 0; i-- {
+					act := activities[i]
+					t.Logf("EC activity: stage=%v message=%v occurred_at=%v", act["stage"], act["message"], act["occurred_at"])
+				}
+			}
 		}
-		t.Logf("Waiting for EC... (Workers: %d, Active Tasks: %d)", workerCount, taskCount)
 
 		time.Sleep(10 * time.Second)
 	}

test/foundationdb/Dockerfile.build

@@ -1,6 +1,6 @@
 # Simplified single-stage build for SeaweedFS with FoundationDB support
 # Force x86_64 platform to use AMD64 FoundationDB packages
-FROM --platform=linux/amd64 golang:1.24-bookworm
+FROM --platform=linux/amd64 golang:1.25-bookworm
 
 ARG FOUNDATIONDB_VERSION=7.4.5
 ENV FOUNDATIONDB_VERSION=${FOUNDATIONDB_VERSION}

test/foundationdb/Dockerfile.build.arm64

@@ -1,5 +1,5 @@
 # Multi-stage Dockerfile to build SeaweedFS with FoundationDB support for ARM64
-FROM --platform=linux/arm64 golang:1.24-bookworm AS builder
+FROM --platform=linux/arm64 golang:1.25-bookworm AS builder
 
 ARG FOUNDATIONDB_VERSION=7.4.5
 ENV FOUNDATIONDB_VERSION=${FOUNDATIONDB_VERSION}

test/foundationdb/Dockerfile.test

@@ -1,5 +1,5 @@
 # Test environment with Go and FoundationDB support
-FROM golang:1.24-bookworm
+FROM golang:1.25-bookworm
 
 # Install system dependencies
 RUN apt-get update && apt-get install -y \

test/java/spark/pom.xml

@@ -123,7 +123,7 @@
             <dependency>
                 <groupId>org.apache.zookeeper</groupId>
                 <artifactId>zookeeper</artifactId>
-                <version>3.9.4</version>
+                <version>3.9.5</version>
             </dependency>
 
             <!-- Apache Commons - Fix CVEs -->

test/kafka/Dockerfile.kafka-gateway

@@ -1,5 +1,5 @@
 # Dockerfile for Kafka Gateway Integration Testing
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git make gcc musl-dev sqlite-dev

test/kafka/Dockerfile.seaweedfs

@@ -1,5 +1,5 @@
 # Dockerfile for building SeaweedFS components from the current workspace
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 RUN apk add --no-cache git make gcc musl-dev sqlite-dev
 

test/kafka/Dockerfile.test-setup

@@ -1,5 +1,5 @@
 # Dockerfile for Kafka Integration Test Setup
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git make gcc musl-dev

test/kafka/kafka-client-loadtest/Dockerfile.loadtest

@@ -2,7 +2,7 @@
 # Multi-stage build for cross-platform support
 
 # Stage 1: Builder
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 WORKDIR /app
 

test/plugin_workers/fake_volume_server.go

@@ -27,19 +27,21 @@ type VolumeServer struct {
 	address  string
 	baseDir  string
 
-	mu                 sync.Mutex
-	receivedFiles      map[string]uint64
-	mountRequests      []*volume_server_pb.VolumeEcShardsMountRequest
-	deleteRequests     []*volume_server_pb.VolumeDeleteRequest
-	markReadonlyCalls  int
-	vacuumGarbageRatio float64
-	vacuumCheckCalls   int
-	vacuumCompactCalls int
-	vacuumCommitCalls  int
-	vacuumCleanupCalls int
-	volumeCopyCalls    int
-	volumeMountCalls   int
-	tailReceiverCalls  int
+	mu                  sync.Mutex
+	receivedFiles       map[string]uint64
+	mountRequests       []*volume_server_pb.VolumeEcShardsMountRequest
+	deleteRequests      []*volume_server_pb.VolumeDeleteRequest
+	markReadonlyCalls   int
+	markWritableCalls   int
+	readFileStatusCalls int
+	vacuumGarbageRatio  float64
+	vacuumCheckCalls    int
+	vacuumCompactCalls  int
+	vacuumCommitCalls   int
+	vacuumCleanupCalls  int
+	volumeCopyCalls     int
+	volumeMountCalls    int
+	tailReceiverCalls   int
 }
 
 // NewVolumeServer starts a test volume server using the provided base directory.
@@ -151,6 +153,20 @@ func (v *VolumeServer) MarkReadonlyCount() int {
 	return v.markReadonlyCalls
 }
 
+// MarkWritableCount returns the number of writable calls.
+func (v *VolumeServer) MarkWritableCount() int {
+	v.mu.Lock()
+	defer v.mu.Unlock()
+	return v.markWritableCalls
+}
+
+// ReadFileStatusCount returns the number of ReadVolumeFileStatus calls.
+func (v *VolumeServer) ReadFileStatusCount() int {
+	v.mu.Lock()
+	defer v.mu.Unlock()
+	return v.readFileStatusCalls
+}
+
 // Shutdown stops the volume server.
 func (v *VolumeServer) Shutdown() {
 	if v.server != nil {
@@ -280,6 +296,25 @@ func (v *VolumeServer) VolumeMarkReadonly(ctx context.Context, req *volume_serve
 	return &volume_server_pb.VolumeMarkReadonlyResponse{}, nil
 }
 
+func (v *VolumeServer) VolumeMarkWritable(ctx context.Context, req *volume_server_pb.VolumeMarkWritableRequest) (*volume_server_pb.VolumeMarkWritableResponse, error) {
+	v.mu.Lock()
+	v.markWritableCalls++
+	v.mu.Unlock()
+	return &volume_server_pb.VolumeMarkWritableResponse{}, nil
+}
+
+func (v *VolumeServer) ReadVolumeFileStatus(ctx context.Context, req *volume_server_pb.ReadVolumeFileStatusRequest) (*volume_server_pb.ReadVolumeFileStatusResponse, error) {
+	v.mu.Lock()
+	v.readFileStatusCalls++
+	v.mu.Unlock()
+	return &volume_server_pb.ReadVolumeFileStatusResponse{
+		VolumeId:    req.VolumeId,
+		DatFileSize: 1024,
+		IdxFileSize: 16,
+		FileCount:   1,
+	}, nil
+}
+
 func (v *VolumeServer) VacuumVolumeCheck(ctx context.Context, req *volume_server_pb.VacuumVolumeCheckRequest) (*volume_server_pb.VacuumVolumeCheckResponse, error) {
 	v.mu.Lock()
 	v.vacuumCheckCalls++

test/plugin_workers/volume_balance/detection_test.go

@@ -37,7 +37,9 @@ func TestVolumeBalanceDetectionIntegration(t *testing.T) {
 		MasterGrpcAddresses: []string{master.Address()},
 	}, 10)
 	require.NoError(t, err)
-	require.Len(t, proposals, 1)
+	// With default batch_size=20 and 10 overloaded volumes vs 1 underloaded,
+	// all moves are grouped into a single batch proposal.
+	require.Len(t, proposals, 1, "expected exactly one batch proposal")
 
 	proposal := proposals[0]
 	require.Equal(t, "volume_balance", proposal.JobType)
@@ -46,8 +48,15 @@ func TestVolumeBalanceDetectionIntegration(t *testing.T) {
 
 	params := &worker_pb.TaskParams{}
 	require.NoError(t, proto.Unmarshal(paramsValue.GetBytesValue(), params))
-	require.NotEmpty(t, params.Sources)
-	require.NotEmpty(t, params.Targets)
+
+	bp := params.GetBalanceParams()
+	require.NotNil(t, bp, "expected BalanceParams in batch proposal")
+	require.Greater(t, len(bp.Moves), 1, "batch proposal should contain multiple moves")
+	for _, move := range bp.Moves {
+		require.NotZero(t, move.VolumeId)
+		require.NotEmpty(t, move.SourceNode)
+		require.NotEmpty(t, move.TargetNode)
+	}
 }
 
 func buildBalanceVolumeListResponse(t *testing.T) *master_pb.VolumeListResponse {

test/plugin_workers/volume_balance/execution_test.go

@@ -8,10 +8,12 @@ import (
 
 	pluginworkers "github.com/seaweedfs/seaweedfs/test/plugin_workers"
 	"github.com/seaweedfs/seaweedfs/weed/pb/plugin_pb"
+	"github.com/seaweedfs/seaweedfs/weed/pb/worker_pb"
 	pluginworker "github.com/seaweedfs/seaweedfs/weed/plugin/worker"
 	"github.com/stretchr/testify/require"
 	"google.golang.org/grpc"
 	"google.golang.org/grpc/credentials/insecure"
+	"google.golang.org/protobuf/proto"
 )
 
 func TestVolumeBalanceExecutionIntegration(t *testing.T) {
@@ -60,8 +62,92 @@ func TestVolumeBalanceExecutionIntegration(t *testing.T) {
 	require.GreaterOrEqual(t, source.MarkReadonlyCount(), 1)
 	require.GreaterOrEqual(t, len(source.DeleteRequests()), 1)
 
-	copyCalls, mountCalls, tailCalls := target.BalanceStats()
+	copyCalls, _, tailCalls := target.BalanceStats()
 	require.GreaterOrEqual(t, copyCalls, 1)
-	require.GreaterOrEqual(t, mountCalls, 1)
 	require.GreaterOrEqual(t, tailCalls, 1)
 }
+
+func TestVolumeBalanceBatchExecutionIntegration(t *testing.T) {
+	dialOption := grpc.WithTransportCredentials(insecure.NewCredentials())
+	handler := pluginworker.NewVolumeBalanceHandler(dialOption)
+	harness := pluginworkers.NewHarness(t, pluginworkers.HarnessConfig{
+		WorkerOptions: pluginworker.WorkerOptions{
+			GrpcDialOption: dialOption,
+		},
+		Handlers: []pluginworker.JobHandler{handler},
+	})
+	harness.WaitForJobType("volume_balance")
+
+	// Create one source and one target fake volume server.
+	source := pluginworkers.NewVolumeServer(t, "")
+	target := pluginworkers.NewVolumeServer(t, "")
+
+	// Build a batch job with 3 volume moves from source → target.
+	volumeIDs := []uint32{401, 402, 403}
+	moves := make([]*worker_pb.BalanceMoveSpec, len(volumeIDs))
+	for i, vid := range volumeIDs {
+		moves[i] = &worker_pb.BalanceMoveSpec{
+			VolumeId:   vid,
+			SourceNode: source.Address(),
+			TargetNode: target.Address(),
+			Collection: "batch-test",
+		}
+	}
+
+	params := &worker_pb.TaskParams{
+		TaskId: "batch-balance-test",
+		TaskParams: &worker_pb.TaskParams_BalanceParams{
+			BalanceParams: &worker_pb.BalanceTaskParams{
+				MaxConcurrentMoves: 2,
+				Moves:              moves,
+			},
+		},
+	}
+	paramBytes, err := proto.Marshal(params)
+	require.NoError(t, err)
+
+	job := &plugin_pb.JobSpec{
+		JobId:   "batch-balance-test",
+		JobType: "volume_balance",
+		Parameters: map[string]*plugin_pb.ConfigValue{
+			"task_params_pb": {
+				Kind: &plugin_pb.ConfigValue_BytesValue{BytesValue: paramBytes},
+			},
+		},
+	}
+
+	ctx, cancel := context.WithTimeout(context.Background(), 30*time.Second)
+	defer cancel()
+
+	result, err := harness.Plugin().ExecuteJob(ctx, job, nil, 1)
+	require.NoError(t, err)
+	require.NotNil(t, result)
+	require.True(t, result.Success, "batch balance job should succeed; result: %+v", result)
+
+	// Each of the 3 moves should have marked the source readonly and deleted.
+	require.Equal(t, len(volumeIDs), source.MarkReadonlyCount(),
+		"each move should mark source volume readonly")
+	require.Equal(t, len(volumeIDs), len(source.DeleteRequests()),
+		"each move should delete the source volume")
+
+	// Verify delete requests reference the expected volume IDs.
+	deletedVols := make(map[uint32]bool)
+	for _, req := range source.DeleteRequests() {
+		deletedVols[req.VolumeId] = true
+	}
+	for _, vid := range volumeIDs {
+		require.True(t, deletedVols[vid], "volume %d should have been deleted from source", vid)
+	}
+
+	// Pre-delete verification should have called ReadVolumeFileStatus on both
+	// source and target for each volume.
+	require.Equal(t, len(volumeIDs), source.ReadFileStatusCount(),
+		"each move should read source volume status before delete")
+	require.Equal(t, len(volumeIDs), target.ReadFileStatusCount(),
+		"each move should read target volume status before delete")
+
+	// Target should have received copy and tail calls for all 3 volumes.
+	copyCalls, _, tailCalls := target.BalanceStats()
+	require.Equal(t, len(volumeIDs), copyCalls, "target should receive one copy per volume")
+	require.Equal(t, len(volumeIDs), tailCalls, "target should receive one tail per volume")
+}

test/postgres/Dockerfile.client

@@ -1,4 +1,4 @@
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Set working directory
 WORKDIR /app

test/postgres/Dockerfile.producer

@@ -1,4 +1,4 @@
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Set working directory
 WORKDIR /app

test/postgres/Dockerfile.seaweedfs

@@ -1,4 +1,4 @@
-FROM golang:1.24-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install git and other build dependencies
 RUN apk add --no-cache git make

test/s3/cors/s3_cors_test.go

@@ -140,7 +140,7 @@ func TestCORSConfigurationManagement(t *testing.T) {
 		Bucket:            aws.String(bucketName),
 		CORSConfiguration: corsConfig,
 	})
-	assert.NoError(t, err, "Should be able to put CORS configuration")
+	require.NoError(t, err, "Should be able to put CORS configuration")
 
 	// Wait for metadata subscription to update cache
 	time.Sleep(50 * time.Millisecond)
@@ -149,9 +149,9 @@ func TestCORSConfigurationManagement(t *testing.T) {
 	getResp, err := client.GetBucketCors(context.TODO(), &s3.GetBucketCorsInput{
 		Bucket: aws.String(bucketName),
 	})
-	assert.NoError(t, err, "Should be able to get CORS configuration")
-	assert.NotNil(t, getResp.CORSRules, "CORS configuration should not be nil")
-	assert.Len(t, getResp.CORSRules, 1, "Should have one CORS rule")
+	require.NoError(t, err, "Should be able to get CORS configuration")
+	require.NotNil(t, getResp.CORSRules, "CORS configuration should not be nil")
+	require.Len(t, getResp.CORSRules, 1, "Should have one CORS rule")
 
 	rule := getResp.CORSRules[0]
 	assert.Equal(t, []string{"*"}, rule.AllowedHeaders, "Allowed headers should match")

test/s3/iam/Dockerfile.s3

@@ -1,5 +1,5 @@
 # Multi-stage build for SeaweedFS S3 with IAM
-FROM golang:1.23-alpine AS builder
+FROM golang:1.25-alpine AS builder
 
 # Install build dependencies
 RUN apk add --no-cache git make curl wget

test/s3/iam/Makefile

@@ -185,6 +185,9 @@ test-context: ## Test only contextual policy enforcement
 test-presigned: ## Test only presigned URL integration
 	go test -v -run TestS3IAMPresignedURLIntegration ./...
 
+test-group: ## Run IAM group management tests
+	go test -v -run "TestIAMGroup" ./...
+
 test-sts: ## Run all STS tests
 	go test -v -run "TestSTS" ./...
 
@@ -263,7 +266,7 @@ docker-build: ## Build custom SeaweedFS image for Docker tests
 
 # All PHONY targets
 .PHONY: test test-quick run-tests setup start-services stop-services wait-for-services clean logs status debug
-.PHONY: test-auth test-policy test-expiration test-multipart test-bucket-policy test-context test-presigned test-sts test-sts-assume-role test-sts-ldap  
+.PHONY: test-auth test-policy test-expiration test-multipart test-bucket-policy test-context test-presigned test-group test-sts test-sts-assume-role test-sts-ldap  
 .PHONY: benchmark ci watch install-deps docker-test docker-up docker-down docker-logs docker-build
 .PHONY: test-distributed test-performance test-stress test-versioning-stress test-keycloak-full test-all-previously-skipped setup-all-tests help-advanced
 

test/s3/iam/s3_iam_group_test.go

@@ -0,0 +1,792 @@
+package iam
+
+import (
+	"encoding/xml"
+	"io"
+	"net/http"
+	"net/url"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/aws/aws-sdk-go/aws"
+	"github.com/aws/aws-sdk-go/aws/awserr"
+	"github.com/aws/aws-sdk-go/aws/credentials"
+	"github.com/aws/aws-sdk-go/aws/session"
+	"github.com/aws/aws-sdk-go/service/iam"
+	"github.com/aws/aws-sdk-go/service/s3"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// TestIAMGroupLifecycle tests the full lifecycle of group management:
+// CreateGroup, GetGroup, ListGroups, DeleteGroup
+func TestIAMGroupLifecycle(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-group-lifecycle"
+
+	t.Run("create_group", func(t *testing.T) {
+		resp, err := iamClient.CreateGroup(&iam.CreateGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		assert.Equal(t, groupName, *resp.Group.GroupName)
+	})
+
+	t.Run("get_group", func(t *testing.T) {
+		resp, err := iamClient.GetGroup(&iam.GetGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		assert.Equal(t, groupName, *resp.Group.GroupName)
+	})
+
+	t.Run("list_groups_contains_created", func(t *testing.T) {
+		resp, err := iamClient.ListGroups(&iam.ListGroupsInput{})
+		require.NoError(t, err)
+		found := false
+		for _, g := range resp.Groups {
+			if *g.GroupName == groupName {
+				found = true
+				break
+			}
+		}
+		assert.True(t, found, "Created group should appear in ListGroups")
+	})
+
+	t.Run("create_duplicate_group_fails", func(t *testing.T) {
+		_, err := iamClient.CreateGroup(&iam.CreateGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		assert.Error(t, err, "Creating a duplicate group should fail")
+	})
+
+	t.Run("delete_group", func(t *testing.T) {
+		_, err := iamClient.DeleteGroup(&iam.DeleteGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+
+		// Verify it's gone
+		resp, err := iamClient.ListGroups(&iam.ListGroupsInput{})
+		require.NoError(t, err)
+		for _, g := range resp.Groups {
+			assert.NotEqual(t, groupName, *g.GroupName,
+				"Deleted group should not appear in ListGroups")
+		}
+	})
+
+	t.Run("delete_nonexistent_group_fails", func(t *testing.T) {
+		_, err := iamClient.DeleteGroup(&iam.DeleteGroupInput{
+			GroupName: aws.String("nonexistent-group-xyz"),
+		})
+		assert.Error(t, err)
+	})
+}
+
+// TestIAMGroupMembership tests adding and removing users from groups
+func TestIAMGroupMembership(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-group-members"
+	userName := "test-user-for-group"
+
+	// Setup: create group and user
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{
+		GroupName: aws.String(groupName),
+	})
+	require.NoError(t, err)
+	defer iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(groupName)})
+
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+	defer iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)})
+
+	t.Run("add_user_to_group", func(t *testing.T) {
+		_, err := iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		})
+		require.NoError(t, err)
+	})
+
+	t.Run("get_group_shows_member", func(t *testing.T) {
+		resp, err := iamClient.GetGroup(&iam.GetGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		found := false
+		for _, u := range resp.Users {
+			if *u.UserName == userName {
+				found = true
+				break
+			}
+		}
+		assert.True(t, found, "Added user should appear in GetGroup members")
+	})
+
+	t.Run("list_groups_for_user", func(t *testing.T) {
+		resp, err := iamClient.ListGroupsForUser(&iam.ListGroupsForUserInput{
+			UserName: aws.String(userName),
+		})
+		require.NoError(t, err)
+		found := false
+		for _, g := range resp.Groups {
+			if *g.GroupName == groupName {
+				found = true
+				break
+			}
+		}
+		assert.True(t, found, "Group should appear in ListGroupsForUser")
+	})
+
+	t.Run("add_duplicate_member_is_idempotent", func(t *testing.T) {
+		_, err := iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		})
+		// Should succeed (idempotent) or return a benign error
+		// AWS IAM allows duplicate add without error
+		assert.NoError(t, err)
+	})
+
+	t.Run("remove_user_from_group", func(t *testing.T) {
+		_, err := iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		})
+		require.NoError(t, err)
+
+		// Verify removal
+		resp, err := iamClient.GetGroup(&iam.GetGroupInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		for _, u := range resp.Users {
+			assert.NotEqual(t, userName, *u.UserName,
+				"Removed user should not appear in group members")
+		}
+	})
+}
+
+// TestIAMGroupPolicyAttachment tests attaching and detaching policies from groups
+func TestIAMGroupPolicyAttachment(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-group-policies"
+	policyName := "test-group-attach-policy"
+	policyDoc := `{"Version":"2012-10-17","Statement":[{"Effect":"Allow","Action":"s3:ListBucket","Resource":"*"}]}`
+
+	// Setup: create group and policy
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{
+		GroupName: aws.String(groupName),
+	})
+	require.NoError(t, err)
+
+	createPolicyResp, err := iamClient.CreatePolicy(&iam.CreatePolicyInput{
+		PolicyName:     aws.String(policyName),
+		PolicyDocument: aws.String(policyDoc),
+	})
+	require.NoError(t, err)
+	policyArn := createPolicyResp.Policy.Arn
+
+	// Cleanup in correct order: detach policy, delete group, delete policy
+	t.Cleanup(func() {
+		if _, err := iamClient.DetachGroupPolicy(&iam.DetachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: policyArn,
+		}); err != nil {
+			t.Logf("cleanup: failed to detach group policy: %v", err)
+		}
+		if _, err := iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(groupName)}); err != nil {
+			t.Logf("cleanup: failed to delete group: %v", err)
+		}
+		if _, err := iamClient.DeletePolicy(&iam.DeletePolicyInput{PolicyArn: policyArn}); err != nil {
+			t.Logf("cleanup: failed to delete policy: %v", err)
+		}
+	})
+
+	t.Run("attach_group_policy", func(t *testing.T) {
+		_, err := iamClient.AttachGroupPolicy(&iam.AttachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: policyArn,
+		})
+		require.NoError(t, err)
+	})
+
+	t.Run("list_attached_group_policies", func(t *testing.T) {
+		resp, err := iamClient.ListAttachedGroupPolicies(&iam.ListAttachedGroupPoliciesInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		found := false
+		for _, p := range resp.AttachedPolicies {
+			if *p.PolicyName == policyName {
+				found = true
+				break
+			}
+		}
+		assert.True(t, found, "Attached policy should appear in ListAttachedGroupPolicies")
+	})
+
+	t.Run("detach_group_policy", func(t *testing.T) {
+		_, err := iamClient.DetachGroupPolicy(&iam.DetachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: policyArn,
+		})
+		require.NoError(t, err)
+
+		// Verify detachment
+		resp, err := iamClient.ListAttachedGroupPolicies(&iam.ListAttachedGroupPoliciesInput{
+			GroupName: aws.String(groupName),
+		})
+		require.NoError(t, err)
+		for _, p := range resp.AttachedPolicies {
+			assert.NotEqual(t, policyName, *p.PolicyName,
+				"Detached policy should not appear in ListAttachedGroupPolicies")
+		}
+	})
+}
+
+// TestIAMGroupPolicyEnforcement tests that group policies are enforced during S3 operations.
+// Creates a user with no direct policies, adds them to a group with S3 access,
+// and verifies they can access S3 through the group policy.
+func TestIAMGroupPolicyEnforcement(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-enforcement-group"
+	userName := "test-enforcement-user"
+	policyName := "test-enforcement-policy"
+	bucketName := "test-group-enforce-bucket"
+	policyDoc := `{"Version":"2012-10-17","Statement":[{"Effect":"Allow","Action":["s3:*"],"Resource":["arn:aws:s3:::` + bucketName + `","arn:aws:s3:::` + bucketName + `/*"]}]}`
+
+	// Create user
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	// Create access key for the user
+	keyResp, err := iamClient.CreateAccessKey(&iam.CreateAccessKeyInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	accessKeyId := *keyResp.AccessKey.AccessKeyId
+	secretKey := *keyResp.AccessKey.SecretAccessKey
+
+	// Create an S3 client with the user's credentials
+	userS3Client := createS3Client(t, accessKeyId, secretKey)
+
+	// Create group
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{
+		GroupName: aws.String(groupName),
+	})
+	require.NoError(t, err)
+
+	// Create policy
+	createPolicyResp, err := iamClient.CreatePolicy(&iam.CreatePolicyInput{
+		PolicyName:     aws.String(policyName),
+		PolicyDocument: aws.String(policyDoc),
+	})
+	require.NoError(t, err)
+	policyArn := createPolicyResp.Policy.Arn
+
+	// Cleanup in correct order: remove user from group, detach policy,
+	// delete access key, delete user, delete group, delete policy
+	t.Cleanup(func() {
+		if _, err := iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		}); err != nil {
+			t.Logf("cleanup: failed to remove user from group: %v", err)
+		}
+		if _, err := iamClient.DetachGroupPolicy(&iam.DetachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: policyArn,
+		}); err != nil {
+			t.Logf("cleanup: failed to detach group policy: %v", err)
+		}
+		if _, err := iamClient.DeleteAccessKey(&iam.DeleteAccessKeyInput{
+			UserName:    aws.String(userName),
+			AccessKeyId: keyResp.AccessKey.AccessKeyId,
+		}); err != nil {
+			t.Logf("cleanup: failed to delete access key: %v", err)
+		}
+		if _, err := iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)}); err != nil {
+			t.Logf("cleanup: failed to delete user: %v", err)
+		}
+		if _, err := iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(groupName)}); err != nil {
+			t.Logf("cleanup: failed to delete group: %v", err)
+		}
+		if _, err := iamClient.DeletePolicy(&iam.DeletePolicyInput{PolicyArn: policyArn}); err != nil {
+			t.Logf("cleanup: failed to delete policy: %v", err)
+		}
+	})
+
+	// Register bucket cleanup on parent test with admin credentials
+	// (userS3Client may lack permissions by cleanup time)
+	adminS3, err := framework.CreateS3ClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+	t.Cleanup(func() {
+		if _, err := adminS3.DeleteObject(&s3.DeleteObjectInput{
+			Bucket: aws.String(bucketName),
+			Key:    aws.String("test-key"),
+		}); err != nil {
+			t.Logf("cleanup: failed to delete object: %v", err)
+		}
+		if _, err := adminS3.DeleteBucket(&s3.DeleteBucketInput{Bucket: aws.String(bucketName)}); err != nil {
+			t.Logf("cleanup: failed to delete bucket: %v", err)
+		}
+	})
+
+	t.Run("user_without_group_denied", func(t *testing.T) {
+		// User has no policies and is not in any group — should be denied
+		_, err := userS3Client.CreateBucket(&s3.CreateBucketInput{
+			Bucket: aws.String(bucketName),
+		})
+		require.Error(t, err, "User without any policies should be denied")
+		awsErr, ok := err.(awserr.Error)
+		require.True(t, ok, "Expected awserr.Error")
+		assert.Equal(t, "AccessDenied", awsErr.Code())
+	})
+
+	t.Run("user_with_group_policy_allowed", func(t *testing.T) {
+		// Attach policy to group
+		_, err := iamClient.AttachGroupPolicy(&iam.AttachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: policyArn,
+		})
+		require.NoError(t, err)
+
+		// Add user to group
+		_, err = iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		})
+		require.NoError(t, err)
+
+		// Wait for policy propagation, then create bucket
+		require.Eventually(t, func() bool {
+			_, err = userS3Client.CreateBucket(&s3.CreateBucketInput{
+				Bucket: aws.String(bucketName),
+			})
+			return err == nil
+		}, 10*time.Second, 500*time.Millisecond, "User with group policy should be allowed")
+
+		// Should also be able to put/get objects
+		_, err = userS3Client.PutObject(&s3.PutObjectInput{
+			Bucket: aws.String(bucketName),
+			Key:    aws.String("test-key"),
+			Body:   aws.ReadSeekCloser(strings.NewReader("test-data")),
+		})
+		require.NoError(t, err, "User should be able to put objects through group policy")
+	})
+
+	t.Run("user_removed_from_group_denied", func(t *testing.T) {
+		// Remove user from group
+		_, err := iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(groupName),
+			UserName:  aws.String(userName),
+		})
+		require.NoError(t, err)
+
+		// Wait for policy propagation — user should now be denied
+		var lastErr error
+		require.Eventually(t, func() bool {
+			_, lastErr = userS3Client.ListObjects(&s3.ListObjectsInput{
+				Bucket: aws.String(bucketName),
+			})
+			return lastErr != nil
+		}, 10*time.Second, 500*time.Millisecond, "User removed from group should be denied")
+		awsErr, ok := lastErr.(awserr.Error)
+		require.True(t, ok, "Expected awserr.Error")
+		assert.Equal(t, "AccessDenied", awsErr.Code())
+	})
+}
+
+// TestIAMGroupDisabledPolicyEnforcement tests that disabled groups do not contribute policies.
+// Uses the raw IAM API (callIAMAPI) since the AWS SDK doesn't support custom group status.
+func TestIAMGroupDisabledPolicyEnforcement(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+	if !isSeaweedFSRunning(t) {
+		t.Skip("SeaweedFS is not running at", TestIAMEndpoint)
+	}
+
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-disabled-group"
+	userName := "test-disabled-grp-user"
+	policyName := "test-disabled-grp-policy"
+	bucketName := "test-disabled-grp-bucket"
+	policyDoc := `{"Version":"2012-10-17","Statement":[{"Effect":"Allow","Action":["s3:*"],"Resource":["arn:aws:s3:::` + bucketName + `","arn:aws:s3:::` + bucketName + `/*"]}]}`
+
+	// Create user, group, policy
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+	require.NoError(t, err)
+
+	keyResp, err := iamClient.CreateAccessKey(&iam.CreateAccessKeyInput{UserName: aws.String(userName)})
+	require.NoError(t, err)
+
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{GroupName: aws.String(groupName)})
+	require.NoError(t, err)
+
+	createPolicyResp, err := iamClient.CreatePolicy(&iam.CreatePolicyInput{
+		PolicyName: aws.String(policyName), PolicyDocument: aws.String(policyDoc),
+	})
+	require.NoError(t, err)
+
+	// Cleanup in correct order: remove user from group, detach policy,
+	// delete access key, delete user, delete group, delete policy
+	t.Cleanup(func() {
+		if _, err := iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(groupName), UserName: aws.String(userName),
+		}); err != nil {
+			t.Logf("cleanup: failed to remove user from group: %v", err)
+		}
+		if _, err := iamClient.DetachGroupPolicy(&iam.DetachGroupPolicyInput{
+			GroupName: aws.String(groupName),
+			PolicyArn: aws.String("arn:aws:iam:::policy/" + policyName),
+		}); err != nil {
+			t.Logf("cleanup: failed to detach group policy: %v", err)
+		}
+		if _, err := iamClient.DeleteAccessKey(&iam.DeleteAccessKeyInput{
+			UserName: aws.String(userName), AccessKeyId: keyResp.AccessKey.AccessKeyId,
+		}); err != nil {
+			t.Logf("cleanup: failed to delete access key: %v", err)
+		}
+		if _, err := iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)}); err != nil {
+			t.Logf("cleanup: failed to delete user: %v", err)
+		}
+		if _, err := iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(groupName)}); err != nil {
+			t.Logf("cleanup: failed to delete group: %v", err)
+		}
+		if _, err := iamClient.DeletePolicy(&iam.DeletePolicyInput{PolicyArn: createPolicyResp.Policy.Arn}); err != nil {
+			t.Logf("cleanup: failed to delete policy: %v", err)
+		}
+	})
+
+	// Setup: attach policy, add user, create bucket with admin
+	_, err = iamClient.AttachGroupPolicy(&iam.AttachGroupPolicyInput{
+		GroupName: aws.String(groupName), PolicyArn: createPolicyResp.Policy.Arn,
+	})
+	require.NoError(t, err)
+
+	_, err = iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+		GroupName: aws.String(groupName), UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	userS3Client := createS3Client(t, *keyResp.AccessKey.AccessKeyId, *keyResp.AccessKey.SecretAccessKey)
+
+	// Create bucket using admin first so we can test listing
+	adminS3, err := framework.CreateS3ClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+	_, err = adminS3.CreateBucket(&s3.CreateBucketInput{Bucket: aws.String(bucketName)})
+	require.NoError(t, err)
+	defer adminS3.DeleteBucket(&s3.DeleteBucketInput{Bucket: aws.String(bucketName)})
+
+	t.Run("enabled_group_allows_access", func(t *testing.T) {
+		require.Eventually(t, func() bool {
+			_, err := userS3Client.ListObjects(&s3.ListObjectsInput{
+				Bucket: aws.String(bucketName),
+			})
+			return err == nil
+		}, 10*time.Second, 500*time.Millisecond, "User in enabled group should have access")
+	})
+
+	t.Run("disabled_group_denies_access", func(t *testing.T) {
+		// Disable group via raw IAM API (no SDK support for this extension)
+		resp, err := callIAMAPIAuthenticated(t, framework, "UpdateGroup", url.Values{
+			"GroupName": {groupName},
+			"Disabled":  {"true"},
+		})
+		require.NoError(t, err)
+		defer resp.Body.Close()
+		require.Equal(t, http.StatusOK, resp.StatusCode, "UpdateGroup (disable) should return 200")
+
+		// Wait for propagation — user should be denied
+		var lastErr error
+		require.Eventually(t, func() bool {
+			_, lastErr = userS3Client.ListObjects(&s3.ListObjectsInput{
+				Bucket: aws.String(bucketName),
+			})
+			return lastErr != nil
+		}, 10*time.Second, 500*time.Millisecond, "User in disabled group should be denied access")
+		awsErr, ok := lastErr.(awserr.Error)
+		require.True(t, ok, "Expected awserr.Error")
+		assert.Equal(t, "AccessDenied", awsErr.Code())
+	})
+
+	t.Run("re_enabled_group_restores_access", func(t *testing.T) {
+		// Re-enable the group
+		resp, err := callIAMAPIAuthenticated(t, framework, "UpdateGroup", url.Values{
+			"GroupName": {groupName},
+			"Disabled":  {"false"},
+		})
+		require.NoError(t, err)
+		defer resp.Body.Close()
+		require.Equal(t, http.StatusOK, resp.StatusCode, "UpdateGroup (re-enable) should return 200")
+
+		// Wait for propagation — user should have access again
+		require.Eventually(t, func() bool {
+			_, err = userS3Client.ListObjects(&s3.ListObjectsInput{
+				Bucket: aws.String(bucketName),
+			})
+			return err == nil
+		}, 10*time.Second, 500*time.Millisecond, "User in re-enabled group should have access again")
+	})
+}
+
+// TestIAMGroupUserDeletionSideEffect tests that deleting a user removes them from all groups.
+func TestIAMGroupUserDeletionSideEffect(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	groupName := "test-deletion-group"
+	userName := "test-deletion-user"
+
+	// Create group and user
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{GroupName: aws.String(groupName)})
+	require.NoError(t, err)
+	defer iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(groupName)})
+
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+	require.NoError(t, err)
+	t.Cleanup(func() {
+		// Best-effort: user may already be deleted by the test
+		iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)})
+	})
+
+	// Add user to group
+	_, err = iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+		GroupName: aws.String(groupName),
+		UserName:  aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	// Verify user is in group
+	getResp, err := iamClient.GetGroup(&iam.GetGroupInput{GroupName: aws.String(groupName)})
+	require.NoError(t, err)
+	assert.Len(t, getResp.Users, 1, "Group should have 1 member before deletion")
+
+	// Delete the user
+	_, err = iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)})
+	require.NoError(t, err)
+
+	// Verify user was removed from the group
+	getResp, err = iamClient.GetGroup(&iam.GetGroupInput{GroupName: aws.String(groupName)})
+	require.NoError(t, err)
+	assert.Empty(t, getResp.Users, "Group should have no members after user deletion")
+}
+
+// TestIAMGroupMultipleGroups tests that a user can belong to multiple groups
+// and inherits policies from all of them.
+func TestIAMGroupMultipleGroups(t *testing.T) {
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	iamClient, err := framework.CreateIAMClientWithJWT("admin-user", "TestAdminRole")
+	require.NoError(t, err)
+
+	group1 := "test-multi-group-1"
+	group2 := "test-multi-group-2"
+	userName := "test-multi-group-user"
+
+	// Create two groups
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{GroupName: aws.String(group1)})
+	require.NoError(t, err)
+	defer iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(group1)})
+
+	_, err = iamClient.CreateGroup(&iam.CreateGroupInput{GroupName: aws.String(group2)})
+	require.NoError(t, err)
+	defer iamClient.DeleteGroup(&iam.DeleteGroupInput{GroupName: aws.String(group2)})
+
+	// Create user
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+	require.NoError(t, err)
+	defer func() {
+		iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(group1), UserName: aws.String(userName),
+		})
+		iamClient.RemoveUserFromGroup(&iam.RemoveUserFromGroupInput{
+			GroupName: aws.String(group2), UserName: aws.String(userName),
+		})
+		iamClient.DeleteUser(&iam.DeleteUserInput{UserName: aws.String(userName)})
+	}()
+
+	// Add user to both groups
+	_, err = iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+		GroupName: aws.String(group1), UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	_, err = iamClient.AddUserToGroup(&iam.AddUserToGroupInput{
+		GroupName: aws.String(group2), UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+
+	// Verify user appears in both groups
+	resp, err := iamClient.ListGroupsForUser(&iam.ListGroupsForUserInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+	groupNames := make(map[string]bool)
+	for _, g := range resp.Groups {
+		groupNames[*g.GroupName] = true
+	}
+	assert.True(t, groupNames[group1], "User should be in group 1")
+	assert.True(t, groupNames[group2], "User should be in group 2")
+}
+
+// --- Response types for raw IAM API calls ---
+
+type CreateGroupResponse struct {
+	XMLName           xml.Name `xml:"CreateGroupResponse"`
+	CreateGroupResult struct {
+		Group struct {
+			GroupName string `xml:"GroupName"`
+		} `xml:"Group"`
+	} `xml:"CreateGroupResult"`
+}
+
+type ListGroupsResponse struct {
+	XMLName          xml.Name `xml:"ListGroupsResponse"`
+	ListGroupsResult struct {
+		Groups []struct {
+			GroupName string `xml:"GroupName"`
+		} `xml:"Groups>member"`
+	} `xml:"ListGroupsResult"`
+}
+
+// callIAMAPIAuthenticated sends an authenticated raw IAM API request using the
+// framework's JWT token. This is needed for custom extensions not in the AWS SDK
+// (like UpdateGroup with Disabled parameter).
+func callIAMAPIAuthenticated(_ *testing.T, framework *S3IAMTestFramework, action string, params url.Values) (*http.Response, error) {
+	params.Set("Action", action)
+
+	req, err := http.NewRequest(http.MethodPost, TestIAMEndpoint+"/",
+		strings.NewReader(params.Encode()))
+	if err != nil {
+		return nil, err
+	}
+
+	req.Header.Set("Content-Type", "application/x-www-form-urlencoded")
+
+	token, err := framework.generateSTSSessionToken("admin-user", "TestAdminRole", time.Hour, "", nil)
+	if err != nil {
+		return nil, err
+	}
+
+	client := &http.Client{
+		Timeout:   30 * time.Second,
+		Transport: &BearerTokenTransport{Token: token},
+	}
+	return client.Do(req)
+}
+
+// TestIAMGroupRawAPI tests group operations using raw HTTP IAM API calls,
+// verifying XML response format for group operations.
+func TestIAMGroupRawAPI(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+	if !isSeaweedFSRunning(t) {
+		t.Skip("SeaweedFS is not running at", TestIAMEndpoint)
+	}
+
+	framework := NewS3IAMTestFramework(t)
+	defer framework.Cleanup()
+
+	groupName := "test-raw-api-group"
+
+	t.Run("create_group_raw", func(t *testing.T) {
+		resp, err := callIAMAPIAuthenticated(t, framework, "CreateGroup", url.Values{
+			"GroupName": {groupName},
+		})
+		require.NoError(t, err)
+		defer resp.Body.Close()
+		assert.Equal(t, http.StatusOK, resp.StatusCode)
+
+		body, err := io.ReadAll(resp.Body)
+		require.NoError(t, err)
+
+		var createResp CreateGroupResponse
+		err = xml.Unmarshal(body, &createResp)
+		require.NoError(t, err)
+		assert.Equal(t, groupName, createResp.CreateGroupResult.Group.GroupName)
+	})
+
+	t.Run("list_groups_raw", func(t *testing.T) {
+		resp, err := callIAMAPIAuthenticated(t, framework, "ListGroups", url.Values{})
+		require.NoError(t, err)
+		defer resp.Body.Close()
+		assert.Equal(t, http.StatusOK, resp.StatusCode)
+
+		body, err := io.ReadAll(resp.Body)
+		require.NoError(t, err)
+
+		var listResp ListGroupsResponse
+		err = xml.Unmarshal(body, &listResp)
+		require.NoError(t, err)
+
+		found := false
+		for _, g := range listResp.ListGroupsResult.Groups {
+			if g.GroupName == groupName {
+				found = true
+				break
+			}
+		}
+		assert.True(t, found, "Created group should appear in raw ListGroups")
+	})
+
+	t.Run("delete_group_raw", func(t *testing.T) {
+		resp, err := callIAMAPIAuthenticated(t, framework, "DeleteGroup", url.Values{
+			"GroupName": {groupName},
+		})
+		require.NoError(t, err)
+		defer resp.Body.Close()
+		assert.Equal(t, http.StatusOK, resp.StatusCode)
+	})
+}
+
+// createS3Client creates an S3 client with static credentials
+func createS3Client(t *testing.T, accessKey, secretKey string) *s3.S3 {
+	sess, err := session.NewSession(&aws.Config{
+		Region:           aws.String("us-east-1"),
+		Endpoint:         aws.String(TestS3Endpoint),
+		Credentials:      credentials.NewStaticCredentials(accessKey, secretKey, ""),
+		DisableSSL:       aws.Bool(true),
+		S3ForcePathStyle: aws.Bool(true),
+	})
+	require.NoError(t, err)
+	return s3.New(sess)
+}

test/s3/normal/get_object_attributes_test.go

@@ -0,0 +1,573 @@
+package example
+
+import (
+	"bytes"
+	"context"
+	"fmt"
+	"io"
+	"net/http"
+	"strings"
+	"testing"
+	"time"
+
+	"github.com/aws/aws-sdk-go/aws"
+	v1credentials "github.com/aws/aws-sdk-go/aws/credentials"
+	v1signer "github.com/aws/aws-sdk-go/aws/signer/v4"
+	v1s3 "github.com/aws/aws-sdk-go/service/s3"
+	v2aws "github.com/aws/aws-sdk-go-v2/aws"
+	"github.com/aws/aws-sdk-go-v2/credentials"
+	v2s3 "github.com/aws/aws-sdk-go-v2/service/s3"
+	"github.com/aws/aws-sdk-go-v2/service/s3/types"
+	"github.com/stretchr/testify/assert"
+	"github.com/stretchr/testify/require"
+)
+
+// newS3V2Client creates an AWS SDK v2 S3 client from the test cluster.
+func newS3V2Client(cluster *TestCluster) *v2s3.Client {
+	return v2s3.New(v2s3.Options{
+		Region:       testRegion,
+		BaseEndpoint: v2aws.String(cluster.s3Endpoint),
+		Credentials:  v2aws.NewCredentialsCache(credentials.NewStaticCredentialsProvider(testAccessKey, testSecretKey, "")),
+		UsePathStyle: true,
+	})
+}
+
+func TestGetObjectAttributes(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+
+	cluster, err := startMiniCluster(t)
+	require.NoError(t, err)
+	defer cluster.Stop()
+
+	t.Run("Basic", func(t *testing.T) {
+		testGetObjectAttributesBasic(t, cluster)
+	})
+	t.Run("MultipartObject", func(t *testing.T) {
+		testGetObjectAttributesMultipart(t, cluster)
+	})
+	t.Run("SelectiveAttributes", func(t *testing.T) {
+		testGetObjectAttributesSelective(t, cluster)
+	})
+	t.Run("InvalidAttribute", func(t *testing.T) {
+		testGetObjectAttributesInvalid(t, cluster)
+	})
+	t.Run("NonExistentObject", func(t *testing.T) {
+		testGetObjectAttributesNotFound(t, cluster)
+	})
+	t.Run("VersionedObject", func(t *testing.T) {
+		testGetObjectAttributesVersioned(t, cluster)
+	})
+	t.Run("ConditionalHeaders", func(t *testing.T) {
+		testGetObjectAttributesConditionalHeaders(t, cluster)
+	})
+	t.Run("VersionedConditionalHeaders", func(t *testing.T) {
+		testGetObjectAttributesVersionedConditionalHeaders(t, cluster)
+	})
+}
+
+func testGetObjectAttributesBasic(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-basic-")
+	objectKey := "test-object.txt"
+	objectData := "Hello, GetObjectAttributes!"
+
+	_, err := cluster.s3Client.PutObject(&v1s3.PutObjectInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+		Body:   bytes.NewReader([]byte(objectData)),
+	})
+	require.NoError(t, err)
+
+	client := newS3V2Client(cluster)
+	resp, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String(objectKey),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesEtag,
+			types.ObjectAttributesStorageClass,
+			types.ObjectAttributesObjectSize,
+			types.ObjectAttributesObjectParts,
+		},
+	})
+	require.NoError(t, err)
+
+	// ETag should be present and non-empty
+	require.NotNil(t, resp.ETag)
+	assert.NotEmpty(t, *resp.ETag)
+	assert.False(t, strings.Contains(*resp.ETag, `"`), "ETag in XML body should not have quotes")
+
+	// ObjectSize should match
+	require.NotNil(t, resp.ObjectSize)
+	assert.Equal(t, int64(len(objectData)), *resp.ObjectSize)
+
+	// StorageClass should be STANDARD (default)
+	assert.Equal(t, "STANDARD", string(resp.StorageClass))
+
+	// ObjectParts should be nil for non-multipart objects
+	assert.Nil(t, resp.ObjectParts)
+
+	// LastModified header should be present
+	assert.NotNil(t, resp.LastModified)
+
+	t.Logf("Basic GetObjectAttributes passed: ETag=%s, Size=%d, StorageClass=%s",
+		*resp.ETag, *resp.ObjectSize, resp.StorageClass)
+}
+
+func testGetObjectAttributesMultipart(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-mp-")
+	objectKey := "test-multipart.bin"
+
+	// Create a 2-part multipart upload
+	part1Data := bytes.Repeat([]byte("A"), 5*1024*1024) // 5MB (minimum part size)
+	part2Data := bytes.Repeat([]byte("B"), 3*1024*1024) // 3MB
+
+	initResp, err := cluster.s3Client.CreateMultipartUpload(&v1s3.CreateMultipartUploadInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+	})
+	require.NoError(t, err)
+	uploadID := initResp.UploadId
+
+	part1Resp, err := cluster.s3Client.UploadPart(&v1s3.UploadPartInput{
+		Bucket:     aws.String(bucketName),
+		Key:        aws.String(objectKey),
+		PartNumber: aws.Int64(1),
+		UploadId:   uploadID,
+		Body:       bytes.NewReader(part1Data),
+	})
+	require.NoError(t, err)
+
+	part2Resp, err := cluster.s3Client.UploadPart(&v1s3.UploadPartInput{
+		Bucket:     aws.String(bucketName),
+		Key:        aws.String(objectKey),
+		PartNumber: aws.Int64(2),
+		UploadId:   uploadID,
+		Body:       bytes.NewReader(part2Data),
+	})
+	require.NoError(t, err)
+
+	_, err = cluster.s3Client.CompleteMultipartUpload(&v1s3.CompleteMultipartUploadInput{
+		Bucket:   aws.String(bucketName),
+		Key:      aws.String(objectKey),
+		UploadId: uploadID,
+		MultipartUpload: &v1s3.CompletedMultipartUpload{
+			Parts: []*v1s3.CompletedPart{
+				{ETag: part1Resp.ETag, PartNumber: aws.Int64(1)},
+				{ETag: part2Resp.ETag, PartNumber: aws.Int64(2)},
+			},
+		},
+	})
+	require.NoError(t, err)
+
+	// Wait briefly for metadata to settle
+	time.Sleep(200 * time.Millisecond)
+
+	client := newS3V2Client(cluster)
+	resp, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String(objectKey),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesObjectParts,
+			types.ObjectAttributesObjectSize,
+		},
+	})
+	require.NoError(t, err)
+
+	require.NotNil(t, resp.ObjectSize)
+	assert.Equal(t, int64(len(part1Data)+len(part2Data)), *resp.ObjectSize)
+
+	require.NotNil(t, resp.ObjectParts, "ObjectParts should be present for multipart objects")
+	assert.Equal(t, int32(2), *resp.ObjectParts.TotalPartsCount)
+	require.Len(t, resp.ObjectParts.Parts, 2)
+	assert.Equal(t, int32(1), *resp.ObjectParts.Parts[0].PartNumber)
+	assert.Equal(t, int64(len(part1Data)), *resp.ObjectParts.Parts[0].Size)
+	assert.Equal(t, int32(2), *resp.ObjectParts.Parts[1].PartNumber)
+	assert.Equal(t, int64(len(part2Data)), *resp.ObjectParts.Parts[1].Size)
+
+	// Test pagination: MaxParts=1
+	resp2, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket:  v2aws.String(bucketName),
+		Key:     v2aws.String(objectKey),
+		MaxParts: v2aws.Int32(1),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesObjectParts,
+		},
+	})
+	require.NoError(t, err)
+	require.NotNil(t, resp2.ObjectParts)
+	assert.Len(t, resp2.ObjectParts.Parts, 1)
+	assert.True(t, *resp2.ObjectParts.IsTruncated)
+	assert.Equal(t, int32(2), *resp2.ObjectParts.TotalPartsCount)
+
+	t.Logf("Multipart GetObjectAttributes passed: %d parts, total size %d",
+		*resp.ObjectParts.TotalPartsCount, *resp.ObjectSize)
+}
+
+func testGetObjectAttributesSelective(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-sel-")
+	objectKey := "test-selective.txt"
+	objectData := "Selective attributes test"
+
+	_, err := cluster.s3Client.PutObject(&v1s3.PutObjectInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+		Body:   bytes.NewReader([]byte(objectData)),
+	})
+	require.NoError(t, err)
+
+	client := newS3V2Client(cluster)
+
+	// Request only ETag
+	resp, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String(objectKey),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesEtag,
+		},
+	})
+	require.NoError(t, err)
+	require.NotNil(t, resp.ETag)
+	assert.NotEmpty(t, *resp.ETag)
+	assert.Nil(t, resp.ObjectSize, "ObjectSize should not be present when not requested")
+	assert.Empty(t, string(resp.StorageClass), "StorageClass should not be present when not requested")
+	assert.Nil(t, resp.ObjectParts, "ObjectParts should not be present when not requested")
+
+	t.Logf("Selective GetObjectAttributes passed: ETag=%s", *resp.ETag)
+}
+
+func testGetObjectAttributesInvalid(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-inv-")
+	objectKey := "test-object.txt"
+
+	_, err := cluster.s3Client.PutObject(&v1s3.PutObjectInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+		Body:   bytes.NewReader([]byte("test")),
+	})
+	require.NoError(t, err)
+
+	// Use raw HTTP to send an invalid attribute name since the SDK validates
+	reqURL := fmt.Sprintf("%s/%s/%s?attributes", cluster.s3Endpoint, bucketName, objectKey)
+	req, err := http.NewRequest("GET", reqURL, nil)
+	require.NoError(t, err)
+	req.Header.Set("X-Amz-Object-Attributes", "InvalidAttr")
+
+	signer := v1signer.NewSigner(v1credentials.NewStaticCredentials(testAccessKey, testSecretKey, ""))
+	_, err = signer.Sign(req, nil, "s3", testRegion, time.Now())
+	require.NoError(t, err)
+
+	client := &http.Client{Timeout: 10 * time.Second}
+	resp, err := client.Do(req)
+	require.NoError(t, err)
+	defer resp.Body.Close()
+	io.Copy(io.Discard, resp.Body)
+
+	assert.Equal(t, 400, resp.StatusCode)
+	t.Logf("Invalid attribute test passed: got %d", resp.StatusCode)
+}
+
+func testGetObjectAttributesNotFound(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-nf-")
+
+	client := newS3V2Client(cluster)
+	_, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("nonexistent-key"),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesEtag,
+		},
+	})
+	require.Error(t, err)
+	assert.Contains(t, err.Error(), "NoSuchKey")
+
+	t.Logf("NotFound GetObjectAttributes passed")
+}
+
+func testGetObjectAttributesVersioned(t *testing.T, cluster *TestCluster) {
+	client := newS3V2Client(cluster)
+	bucketName := createTestBucket(t, cluster, "test-goa-ver-")
+
+	// Enable versioning
+	_, err := client.PutBucketVersioning(context.Background(), &v2s3.PutBucketVersioningInput{
+		Bucket: v2aws.String(bucketName),
+		VersioningConfiguration: &types.VersioningConfiguration{
+			Status: types.BucketVersioningStatusEnabled,
+		},
+	})
+	require.NoError(t, err)
+	time.Sleep(200 * time.Millisecond)
+
+	// Put two versions of the same object
+	v1Data := "version 1 content"
+	putResp1, err := client.PutObject(context.Background(), &v2s3.PutObjectInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("versioned-key"),
+		Body:   strings.NewReader(v1Data),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, putResp1.VersionId)
+	versionId1 := *putResp1.VersionId
+
+	v2Data := "version 2 content - longer"
+	putResp2, err := client.PutObject(context.Background(), &v2s3.PutObjectInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("versioned-key"),
+		Body:   strings.NewReader(v2Data),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, putResp2.VersionId)
+	versionId2 := *putResp2.VersionId
+
+	assert.NotEqual(t, versionId1, versionId2, "versions should differ")
+
+	// GetObjectAttributes for latest version (v2)
+	resp, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("versioned-key"),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesObjectSize,
+			types.ObjectAttributesEtag,
+		},
+	})
+	require.NoError(t, err)
+	require.NotNil(t, resp.ObjectSize)
+	assert.Equal(t, int64(len(v2Data)), *resp.ObjectSize)
+	require.NotNil(t, resp.VersionId)
+	assert.Equal(t, versionId2, *resp.VersionId)
+
+	// GetObjectAttributes for specific older version (v1)
+	resp1, err := client.GetObjectAttributes(context.Background(), &v2s3.GetObjectAttributesInput{
+		Bucket:    v2aws.String(bucketName),
+		Key:       v2aws.String("versioned-key"),
+		VersionId: v2aws.String(versionId1),
+		ObjectAttributes: []types.ObjectAttributes{
+			types.ObjectAttributesObjectSize,
+			types.ObjectAttributesEtag,
+		},
+	})
+	require.NoError(t, err)
+	require.NotNil(t, resp1.ObjectSize)
+	assert.Equal(t, int64(len(v1Data)), *resp1.ObjectSize)
+	require.NotNil(t, resp1.VersionId)
+	assert.Equal(t, versionId1, *resp1.VersionId)
+
+	t.Logf("Versioned GetObjectAttributes passed: v1 size=%d (id=%s), v2 size=%d (id=%s)",
+		*resp1.ObjectSize, versionId1, *resp.ObjectSize, versionId2)
+}
+
+// signedGetObjectAttributes creates a signed GET request for ?attributes with custom headers.
+func signedGetObjectAttributes(t *testing.T, cluster *TestCluster, bucketName, objectKey string, extraHeaders map[string]string) *http.Response {
+	reqURL := fmt.Sprintf("%s/%s/%s?attributes", cluster.s3Endpoint, bucketName, objectKey)
+	req, err := http.NewRequest("GET", reqURL, nil)
+	require.NoError(t, err)
+	req.Header.Set("X-Amz-Object-Attributes", "ETag,ObjectSize")
+	for k, v := range extraHeaders {
+		req.Header.Set(k, v)
+	}
+	signer := v1signer.NewSigner(v1credentials.NewStaticCredentials(testAccessKey, testSecretKey, ""))
+	_, err = signer.Sign(req, nil, "s3", testRegion, time.Now())
+	require.NoError(t, err)
+	client := &http.Client{Timeout: 10 * time.Second}
+	resp, err := client.Do(req)
+	require.NoError(t, err)
+	return resp
+}
+
+func testGetObjectAttributesConditionalHeaders(t *testing.T, cluster *TestCluster) {
+	bucketName := createTestBucket(t, cluster, "test-goa-cond-")
+	objectKey := "cond-test.txt"
+
+	_, err := cluster.s3Client.PutObject(&v1s3.PutObjectInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+		Body:   bytes.NewReader([]byte("conditional headers test")),
+	})
+	require.NoError(t, err)
+
+	// Get the ETag and Last-Modified for the object
+	headResp, err := cluster.s3Client.HeadObject(&v1s3.HeadObjectInput{
+		Bucket: aws.String(bucketName),
+		Key:    aws.String(objectKey),
+	})
+	require.NoError(t, err)
+	etag := aws.StringValue(headResp.ETag)
+	lastModified := headResp.LastModified
+	require.NotNil(t, lastModified)
+
+	pastDate := lastModified.Add(-1 * time.Hour).UTC().Format(http.TimeFormat)
+	futureDate := lastModified.Add(1 * time.Hour).UTC().Format(http.TimeFormat)
+
+	// RFC 7232: If-Match true + If-Unmodified-Since false => 200 OK
+	// If-Unmodified-Since is ignored when If-Match is present
+	t.Run("IfMatch_true_IfUnmodifiedSince_false", func(t *testing.T) {
+		resp := signedGetObjectAttributes(t, cluster, bucketName, objectKey, map[string]string{
+			"If-Match":            etag,
+			"If-Unmodified-Since": pastDate, // object was modified after this => false
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 200, resp.StatusCode,
+			"If-Match=true should return 200 even when If-Unmodified-Since=false (RFC 7232 Section 3.4)")
+	})
+
+	// RFC 7232: If-None-Match false + If-Modified-Since true => 304 Not Modified
+	// If-Modified-Since is ignored when If-None-Match is present
+	t.Run("IfNoneMatch_false_IfModifiedSince_true", func(t *testing.T) {
+		resp := signedGetObjectAttributes(t, cluster, bucketName, objectKey, map[string]string{
+			"If-None-Match":     etag,
+			"If-Modified-Since": pastDate, // object was modified after this => true
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 304, resp.StatusCode,
+			"If-None-Match=false (ETag match) should return 304 even when If-Modified-Since=true (RFC 7232 Section 3.3)")
+	})
+
+	// If-Match succeeds, If-Unmodified-Since also succeeds => 200
+	t.Run("IfMatch_true_IfUnmodifiedSince_true", func(t *testing.T) {
+		resp := signedGetObjectAttributes(t, cluster, bucketName, objectKey, map[string]string{
+			"If-Match":            etag,
+			"If-Unmodified-Since": futureDate,
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 200, resp.StatusCode)
+	})
+
+	// If-None-Match passes (ETag differs), If-Modified-Since ignored => 200
+	// Per RFC 7232, If-Modified-Since is ignored when If-None-Match is present
+	t.Run("IfNoneMatch_true_IfModifiedSince_ignored", func(t *testing.T) {
+		resp := signedGetObjectAttributes(t, cluster, bucketName, objectKey, map[string]string{
+			"If-None-Match":     `"nonexistent-etag"`,
+			"If-Modified-Since": futureDate, // would fail alone, but is ignored
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 200, resp.StatusCode,
+			"If-None-Match=true means If-Modified-Since is ignored, should return 200 (RFC 7232 Section 3.3)")
+	})
+
+	// If-Match fails => 412 regardless of If-Unmodified-Since
+	t.Run("IfMatch_false", func(t *testing.T) {
+		resp := signedGetObjectAttributes(t, cluster, bucketName, objectKey, map[string]string{
+			"If-Match":            `"wrong-etag"`,
+			"If-Unmodified-Since": futureDate,
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 412, resp.StatusCode)
+	})
+
+	t.Logf("Conditional headers tests passed")
+}
+
+// signedGetObjectAttributesVersioned creates a signed GET request for ?attributes&versionId=... with custom headers.
+func signedGetObjectAttributesVersioned(t *testing.T, cluster *TestCluster, bucketName, objectKey, versionId string, extraHeaders map[string]string) *http.Response {
+	reqURL := fmt.Sprintf("%s/%s/%s?attributes&versionId=%s", cluster.s3Endpoint, bucketName, objectKey, versionId)
+	req, err := http.NewRequest("GET", reqURL, nil)
+	require.NoError(t, err)
+	req.Header.Set("X-Amz-Object-Attributes", "ETag,ObjectSize")
+	for k, v := range extraHeaders {
+		req.Header.Set(k, v)
+	}
+	signer := v1signer.NewSigner(v1credentials.NewStaticCredentials(testAccessKey, testSecretKey, ""))
+	_, err = signer.Sign(req, nil, "s3", testRegion, time.Now())
+	require.NoError(t, err)
+	client := &http.Client{Timeout: 10 * time.Second}
+	resp, err := client.Do(req)
+	require.NoError(t, err)
+	return resp
+}
+
+func testGetObjectAttributesVersionedConditionalHeaders(t *testing.T, cluster *TestCluster) {
+	client := newS3V2Client(cluster)
+	bucketName := createTestBucket(t, cluster, "test-goa-vcond-")
+
+	// Enable versioning
+	_, err := client.PutBucketVersioning(context.Background(), &v2s3.PutBucketVersioningInput{
+		Bucket: v2aws.String(bucketName),
+		VersioningConfiguration: &types.VersioningConfiguration{
+			Status: types.BucketVersioningStatusEnabled,
+		},
+	})
+	require.NoError(t, err)
+	time.Sleep(200 * time.Millisecond)
+
+	// Put two versions with different content (different ETags)
+	v1Data := "version 1 - original"
+	putResp1, err := client.PutObject(context.Background(), &v2s3.PutObjectInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("vcond-key"),
+		Body:   strings.NewReader(v1Data),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, putResp1.VersionId)
+	vid1 := *putResp1.VersionId
+
+	v2Data := "version 2 - updated content"
+	putResp2, err := client.PutObject(context.Background(), &v2s3.PutObjectInput{
+		Bucket: v2aws.String(bucketName),
+		Key:    v2aws.String("vcond-key"),
+		Body:   strings.NewReader(v2Data),
+	})
+	require.NoError(t, err)
+	require.NotNil(t, putResp2.VersionId)
+	vid2 := *putResp2.VersionId
+
+	// Get ETags for each version
+	headV1, err := client.HeadObject(context.Background(), &v2s3.HeadObjectInput{
+		Bucket:    v2aws.String(bucketName),
+		Key:       v2aws.String("vcond-key"),
+		VersionId: v2aws.String(vid1),
+	})
+	require.NoError(t, err)
+	etagV1 := *headV1.ETag
+
+	headV2, err := client.HeadObject(context.Background(), &v2s3.HeadObjectInput{
+		Bucket:    v2aws.String(bucketName),
+		Key:       v2aws.String("vcond-key"),
+		VersionId: v2aws.String(vid2),
+	})
+	require.NoError(t, err)
+	etagV2 := *headV2.ETag
+	require.NotEqual(t, etagV1, etagV2, "versions should have different ETags")
+
+	// If-Match with v1's ETag + versionId=v1 => 200
+	// Before the fix, this would fail with 412 because conditional headers
+	// were evaluated against the latest version (v2) whose ETag differs
+	t.Run("IfMatch_v1_etag_versionId_v1", func(t *testing.T) {
+		resp := signedGetObjectAttributesVersioned(t, cluster, bucketName, "vcond-key", vid1, map[string]string{
+			"If-Match": etagV1,
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 200, resp.StatusCode,
+			"If-Match with v1 ETag targeting versionId=v1 should return 200")
+	})
+
+	// If-Match with v2's ETag + versionId=v1 => 412
+	// The ETag doesn't match v1, so this should fail
+	t.Run("IfMatch_v2_etag_versionId_v1", func(t *testing.T) {
+		resp := signedGetObjectAttributesVersioned(t, cluster, bucketName, "vcond-key", vid1, map[string]string{
+			"If-Match": etagV2,
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 412, resp.StatusCode,
+			"If-Match with v2 ETag targeting versionId=v1 should return 412")
+	})
+
+	// If-None-Match with v1's ETag + versionId=v1 => 304
+	t.Run("IfNoneMatch_v1_etag_versionId_v1", func(t *testing.T) {
+		resp := signedGetObjectAttributesVersioned(t, cluster, bucketName, "vcond-key", vid1, map[string]string{
+			"If-None-Match": etagV1,
+		})
+		defer resp.Body.Close()
+		io.Copy(io.Discard, resp.Body)
+		assert.Equal(t, 304, resp.StatusCode,
+			"If-None-Match with v1 ETag targeting versionId=v1 should return 304")
+	})
+
+	t.Logf("Versioned conditional headers tests passed: vid1=%s, vid2=%s", vid1, vid2)
+}

test/s3/policy/policy_test.go

@@ -422,6 +422,218 @@ func TestS3MultipartOperationsInheritPutObjectPermissions(t *testing.T) {
 	require.Equal(t, 0, len(listUploadsOut.Uploads))
 }
 
+// TestS3IAMManagedPolicyLifecycle is an end-to-end integration test covering the
+// user-reported use case in https://github.com/seaweedfs/seaweedfs/issues/8506
+// where managed policy operations (GetPolicy, ListPolicies, DeletePolicy,
+// AttachUserPolicy, DetachUserPolicy) returned 500 errors.
+func TestS3IAMManagedPolicyLifecycle(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+
+	cluster, err := startMiniCluster(t)
+	require.NoError(t, err)
+	defer cluster.Stop()
+
+	iamClient := newIAMClient(t, cluster.s3Endpoint)
+
+	// Step 1: Create a user (this already worked per the issue)
+	userName := uniqueName("lifecycle-user")
+	_, err = iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+	require.NoError(t, err, "CreateUser should succeed")
+
+	// Step 2: Create a managed policy via IAM API
+	policyName := uniqueName("lifecycle-policy")
+	policyArn := fmt.Sprintf("arn:aws:iam:::policy/%s", policyName)
+	policyDoc := `{
+		"Version": "2012-10-17",
+		"Statement": [{
+			"Effect": "Allow",
+			"Action": ["s3:GetObject", "s3:PutObject"],
+			"Resource": "arn:aws:s3:::*"
+		}]
+	}`
+	createOut, err := iamClient.CreatePolicy(&iam.CreatePolicyInput{
+		PolicyName:     aws.String(policyName),
+		PolicyDocument: aws.String(policyDoc),
+	})
+	require.NoError(t, err, "CreatePolicy should succeed")
+	require.NotNil(t, createOut.Policy)
+	require.Equal(t, policyName, *createOut.Policy.PolicyName)
+
+	// Step 3: ListPolicies — should include the created policy (was returning 500)
+	listOut, err := iamClient.ListPolicies(&iam.ListPoliciesInput{})
+	require.NoError(t, err, "ListPolicies should succeed (was returning 500)")
+	require.True(t, managedPolicyContains(listOut.Policies, policyName),
+		"ListPolicies should contain the newly created policy")
+
+	// Step 4: GetPolicy by ARN — should return the policy (was returning 500)
+	getOut, err := iamClient.GetPolicy(&iam.GetPolicyInput{PolicyArn: aws.String(policyArn)})
+	require.NoError(t, err, "GetPolicy should succeed (was returning 500)")
+	require.NotNil(t, getOut.Policy)
+	require.Equal(t, policyName, *getOut.Policy.PolicyName)
+	require.Equal(t, policyArn, *getOut.Policy.Arn)
+
+	// Step 5: AttachUserPolicy — should succeed (was returning 500)
+	_, err = iamClient.AttachUserPolicy(&iam.AttachUserPolicyInput{
+		UserName:  aws.String(userName),
+		PolicyArn: aws.String(policyArn),
+	})
+	require.NoError(t, err, "AttachUserPolicy should succeed (was returning 500)")
+
+	// Step 6: ListAttachedUserPolicies — verify the policy is attached
+	attachedOut, err := iamClient.ListAttachedUserPolicies(&iam.ListAttachedUserPoliciesInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err, "ListAttachedUserPolicies should succeed")
+	require.True(t, attachedPolicyContains(attachedOut.AttachedPolicies, policyName),
+		"Policy should appear in user's attached policies")
+
+	// Step 7: Idempotent re-attach should not fail
+	_, err = iamClient.AttachUserPolicy(&iam.AttachUserPolicyInput{
+		UserName:  aws.String(userName),
+		PolicyArn: aws.String(policyArn),
+	})
+	require.NoError(t, err, "Re-attaching same policy should be idempotent")
+
+	// Step 8: DeletePolicy while attached — should fail with DeleteConflict (AWS behavior)
+	_, err = iamClient.DeletePolicy(&iam.DeletePolicyInput{PolicyArn: aws.String(policyArn)})
+	require.Error(t, err, "DeletePolicy should fail while policy is attached")
+	var awsErr awserr.Error
+	require.True(t, errors.As(err, &awsErr))
+	require.Equal(t, iam.ErrCodeDeleteConflictException, awsErr.Code(),
+		"Should return DeleteConflict when deleting attached policy")
+
+	// Step 9: DetachUserPolicy
+	_, err = iamClient.DetachUserPolicy(&iam.DetachUserPolicyInput{
+		UserName:  aws.String(userName),
+		PolicyArn: aws.String(policyArn),
+	})
+	require.NoError(t, err, "DetachUserPolicy should succeed")
+
+	// Verify detached
+	attachedOut, err = iamClient.ListAttachedUserPolicies(&iam.ListAttachedUserPoliciesInput{
+		UserName: aws.String(userName),
+	})
+	require.NoError(t, err)
+	require.False(t, attachedPolicyContains(attachedOut.AttachedPolicies, policyName),
+		"Policy should no longer appear in user's attached policies after detach")
+
+	// Step 10: DeletePolicy — should now succeed (was returning XML parsing error)
+	_, err = iamClient.DeletePolicy(&iam.DeletePolicyInput{PolicyArn: aws.String(policyArn)})
+	require.NoError(t, err, "DeletePolicy should succeed after detach (was returning XML parsing error)")
+
+	// Step 11: Verify the policy is gone
+	listOut, err = iamClient.ListPolicies(&iam.ListPoliciesInput{})
+	require.NoError(t, err)
+	require.False(t, managedPolicyContains(listOut.Policies, policyName),
+		"Deleted policy should not appear in ListPolicies")
+
+	_, err = iamClient.GetPolicy(&iam.GetPolicyInput{PolicyArn: aws.String(policyArn)})
+	require.Error(t, err, "GetPolicy should fail for deleted policy")
+	require.True(t, errors.As(err, &awsErr))
+	require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+}
+
+// TestS3IAMManagedPolicyErrorCases covers error cases from the user-reported issue:
+// invalid ARNs, missing policies, and missing users.
+func TestS3IAMManagedPolicyErrorCases(t *testing.T) {
+	if testing.Short() {
+		t.Skip("Skipping integration test in short mode")
+	}
+
+	cluster, err := startMiniCluster(t)
+	require.NoError(t, err)
+	defer cluster.Stop()
+
+	iamClient := newIAMClient(t, cluster.s3Endpoint)
+
+	t.Run("GetPolicy with nonexistent ARN returns NoSuchEntity", func(t *testing.T) {
+		_, err := iamClient.GetPolicy(&iam.GetPolicyInput{
+			PolicyArn: aws.String("arn:aws:iam:::policy/does-not-exist"),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+
+	t.Run("DeletePolicy with nonexistent ARN returns NoSuchEntity", func(t *testing.T) {
+		_, err := iamClient.DeletePolicy(&iam.DeletePolicyInput{
+			PolicyArn: aws.String("arn:aws:iam:::policy/does-not-exist"),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+
+	t.Run("AttachUserPolicy with nonexistent policy returns NoSuchEntity", func(t *testing.T) {
+		userName := uniqueName("err-user")
+		_, err := iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+		require.NoError(t, err)
+
+		_, err = iamClient.AttachUserPolicy(&iam.AttachUserPolicyInput{
+			UserName:  aws.String(userName),
+			PolicyArn: aws.String("arn:aws:iam:::policy/does-not-exist"),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+
+	t.Run("AttachUserPolicy with nonexistent user returns NoSuchEntity", func(t *testing.T) {
+		policyName := uniqueName("err-policy")
+		_, err := iamClient.CreatePolicy(&iam.CreatePolicyInput{
+			PolicyName:     aws.String(policyName),
+			PolicyDocument: aws.String(`{"Version":"2012-10-17","Statement":[{"Effect":"Allow","Action":"s3:GetObject","Resource":"*"}]}`),
+		})
+		require.NoError(t, err)
+
+		_, err = iamClient.AttachUserPolicy(&iam.AttachUserPolicyInput{
+			UserName:  aws.String("nonexistent-user"),
+			PolicyArn: aws.String(fmt.Sprintf("arn:aws:iam:::policy/%s", policyName)),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+
+	t.Run("DetachUserPolicy that is not attached returns NoSuchEntity", func(t *testing.T) {
+		userName := uniqueName("detach-user")
+		_, err := iamClient.CreateUser(&iam.CreateUserInput{UserName: aws.String(userName)})
+		require.NoError(t, err)
+
+		policyName := uniqueName("detach-policy")
+		_, err = iamClient.CreatePolicy(&iam.CreatePolicyInput{
+			PolicyName:     aws.String(policyName),
+			PolicyDocument: aws.String(`{"Version":"2012-10-17","Statement":[{"Effect":"Allow","Action":"s3:GetObject","Resource":"*"}]}`),
+		})
+		require.NoError(t, err)
+
+		_, err = iamClient.DetachUserPolicy(&iam.DetachUserPolicyInput{
+			UserName:  aws.String(userName),
+			PolicyArn: aws.String(fmt.Sprintf("arn:aws:iam:::policy/%s", policyName)),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+
+	t.Run("ListAttachedUserPolicies for nonexistent user returns NoSuchEntity", func(t *testing.T) {
+		_, err := iamClient.ListAttachedUserPolicies(&iam.ListAttachedUserPoliciesInput{
+			UserName: aws.String("nonexistent-user"),
+		})
+		require.Error(t, err)
+		var awsErr awserr.Error
+		require.True(t, errors.As(err, &awsErr))
+		require.Equal(t, iam.ErrCodeNoSuchEntityException, awsErr.Code())
+	})
+}
+
 func execShell(t *testing.T, weedCmd, master, filer, shellCmd string) string {
 	// weed shell -master=... -filer=...
 	args := []string{"shell", "-master=" + master, "-filer=" + filer}

test/volume_server/http/admin_test.go

@@ -23,8 +23,6 @@ func TestAdminStatusAndHealthz(t *testing.T) {
 	if err != nil {
 		t.Fatalf("create status request: %v", err)
 	}
-	statusReq.Header.Set(request_id.AmzRequestIDHeader, "test-request-id-1")
-
 	statusResp := framework.DoRequest(t, client, statusReq)
 	statusBody := framework.ReadAllAndClose(t, statusResp)
 
@@ -34,8 +32,8 @@ func TestAdminStatusAndHealthz(t *testing.T) {
 	if got := statusResp.Header.Get("Server"); !strings.Contains(got, "SeaweedFS Volume") {
 		t.Fatalf("expected /status Server header to contain SeaweedFS Volume, got %q", got)
 	}
-	if got := statusResp.Header.Get(request_id.AmzRequestIDHeader); got != "test-request-id-1" {
-		t.Fatalf("expected echoed request id, got %q", got)
+	if got := statusResp.Header.Get(request_id.AmzRequestIDHeader); got == "" {
+		t.Fatal("expected server-generated request id in response header")
 	}
 
 	var payload map[string]interface{}
@@ -49,7 +47,6 @@ func TestAdminStatusAndHealthz(t *testing.T) {
 	}
 
 	healthReq := mustNewRequest(t, http.MethodGet, cluster.VolumeAdminURL()+"/healthz")
-	healthReq.Header.Set(request_id.AmzRequestIDHeader, "test-request-id-2")
 	healthResp := framework.DoRequest(t, client, healthReq)
 	_ = framework.ReadAllAndClose(t, healthResp)
 	if healthResp.StatusCode != http.StatusOK {
@@ -58,8 +55,8 @@ func TestAdminStatusAndHealthz(t *testing.T) {
 	if got := healthResp.Header.Get("Server"); !strings.Contains(got, "SeaweedFS Volume") {
 		t.Fatalf("expected /healthz Server header to contain SeaweedFS Volume, got %q", got)
 	}
-	if got := healthResp.Header.Get(request_id.AmzRequestIDHeader); got != "test-request-id-2" {
-		t.Fatalf("expected /healthz echoed request id, got %q", got)
+	if got := healthResp.Header.Get(request_id.AmzRequestIDHeader); got == "" {
+		t.Fatal("expected /healthz server-generated request id in response header")
 	}
 
 	uiResp := framework.DoRequest(t, client, mustNewRequest(t, http.MethodGet, cluster.VolumeAdminURL()+"/ui/index.html"))

weed/admin/dash/admin_data.go

@@ -80,6 +80,11 @@ type AccessKeyInfo struct {
 	CreatedAt time.Time `json:"created_at"`
 }
 
+type CreateAccessKeyRequest struct {
+	AccessKey string `json:"access_key"`
+	SecretKey string `json:"secret_key"`
+}
+
 type UpdateAccessKeyStatusRequest struct {
 	Status string `json:"status" binding:"required"`
 }
@@ -90,6 +95,7 @@ type UserDetails struct {
 	Actions     []string        `json:"actions"`
 	PolicyNames []string        `json:"policy_names"`
 	AccessKeys  []AccessKeyInfo `json:"access_keys"`
+	Groups      []string        `json:"groups"`
 }
 
 type FilerNode struct {

weed/admin/dash/admin_lock_manager.go

@@ -4,12 +4,13 @@ import (
 	"sync"
 	"time"
 
+	"github.com/seaweedfs/seaweedfs/weed/cluster"
 	"github.com/seaweedfs/seaweedfs/weed/wdclient"
 	"github.com/seaweedfs/seaweedfs/weed/wdclient/exclusive_locks"
 )
 
 const (
-	adminLockName       = "shell"
+	adminLockName       = cluster.AdminShellLockName
 	adminLockClientName = "admin-plugin"
 )
 

weed/admin/dash/admin_presence_lock.go

@@ -0,0 +1,61 @@
+package dash
+
+import (
+	"time"
+
+	"github.com/seaweedfs/seaweedfs/weed/cluster"
+	"github.com/seaweedfs/seaweedfs/weed/wdclient"
+	"github.com/seaweedfs/seaweedfs/weed/wdclient/exclusive_locks"
+)
+
+const adminPresenceClientName = "admin-server"
+
+type adminPresenceLock struct {
+	locker *exclusive_locks.ExclusiveLocker
+	stopCh chan struct{}
+}
+
+func newAdminPresenceLock(masterClient *wdclient.MasterClient) *adminPresenceLock {
+	if masterClient == nil {
+		return nil
+	}
+	return &adminPresenceLock{
+		locker: exclusive_locks.NewExclusiveLocker(masterClient, cluster.AdminServerPresenceLockName),
+		stopCh: make(chan struct{}),
+	}
+}
+
+func (l *adminPresenceLock) Start() {
+	if l == nil || l.locker == nil {
+		return
+	}
+	l.locker.SetMessage("admin server connected")
+	go func() {
+		ticker := time.NewTicker(5 * time.Second)
+		defer ticker.Stop()
+		for {
+			if !l.locker.IsLocked() {
+				l.locker.RequestLock(adminPresenceClientName)
+			}
+			select {
+			case <-l.stopCh:
+				return
+			case <-ticker.C:
+			}
+		}
+	}()
+}
+
+func (l *adminPresenceLock) Stop() {
+	if l == nil {
+		return
+	}
+	select {
+	case <-l.stopCh:
+	default:
+		close(l.stopCh)
+	}
+	if l.locker != nil {
+		l.locker.ReleaseLock()
+	}
+}

weed/admin/dash/admin_server.go

@@ -2,7 +2,9 @@ package dash
 
 import (
 	"context"
+	"errors"
 	"fmt"
+	"io"
 	"net/http"
 	"sort"
 	"strings"
@@ -11,6 +13,7 @@ import (
 	"github.com/seaweedfs/seaweedfs/weed/admin/maintenance"
 	adminplugin "github.com/seaweedfs/seaweedfs/weed/admin/plugin"
 	"github.com/seaweedfs/seaweedfs/weed/cluster"
+	clustermaintenance "github.com/seaweedfs/seaweedfs/weed/cluster/maintenance"
 	"github.com/seaweedfs/seaweedfs/weed/credential"
 	"github.com/seaweedfs/seaweedfs/weed/glog"
 	"github.com/seaweedfs/seaweedfs/weed/pb"
@@ -99,6 +102,7 @@ type AdminServer struct {
 	maintenanceManager *maintenance.MaintenanceManager
 	plugin             *adminplugin.Plugin
 	pluginLock         *AdminLockManager
+	adminPresenceLock  *adminPresenceLock
 	expireJobHandler   func(jobID string, reason string) (*adminplugin.TrackedJob, bool, error)
 
 	// Topic retention purger
@@ -137,6 +141,10 @@ func NewAdminServer(masters string, templateFS http.FileSystem, dataDir string,
 	go masterClient.KeepConnectedToMaster(ctx)
 
 	lockManager := NewAdminLockManager(masterClient, adminLockClientName)
+	presenceLock := newAdminPresenceLock(masterClient)
+	if presenceLock != nil {
+		presenceLock.Start()
+	}
 
 	server := &AdminServer{
 		masterClient:                  masterClient,
@@ -150,6 +158,7 @@ func NewAdminServer(masters string, templateFS http.FileSystem, dataDir string,
 		s3TablesManager:               newS3TablesManager(),
 		icebergPort:                   icebergPort,
 		pluginLock:                    lockManager,
+		adminPresenceLock:             presenceLock,
 	}
 
 	// Initialize topic retention purger
@@ -228,22 +237,19 @@ func NewAdminServer(masters string, templateFS http.FileSystem, dataDir string,
 		}()
 	}
 
-	plugin, err := adminplugin.New(adminplugin.Options{
+	pluginOpts := adminplugin.Options{
 		DataDir: dataDir,
 		ClusterContextProvider: func(_ context.Context) (*plugin_pb.ClusterContext, error) {
 			return server.buildDefaultPluginClusterContext(), nil
 		},
-		LockManager: lockManager,
-	})
+		LockManager:            lockManager,
+		ConfigDefaultsProvider: server.enrichConfigDefaults,
+	}
+	plugin, err := adminplugin.New(pluginOpts)
 	if err != nil && dataDir != "" {
 		glog.Warningf("Failed to initialize plugin with dataDir=%q: %v. Falling back to in-memory plugin state.", dataDir, err)
-		plugin, err = adminplugin.New(adminplugin.Options{
-			DataDir: "",
-			ClusterContextProvider: func(_ context.Context) (*plugin_pb.ClusterContext, error) {
-				return server.buildDefaultPluginClusterContext(), nil
-			},
-			LockManager: lockManager,
-		})
+		pluginOpts.DataDir = ""
+		plugin, err = adminplugin.New(pluginOpts)
 	}
 	if err != nil {
 		glog.Errorf("Failed to initialize plugin: %v", err)
@@ -267,6 +273,89 @@ func (s *AdminServer) loadTaskConfigurationsFromPersistence() {
 	configUpdateRegistry.UpdateAllConfigs(s.configPersistence)
 }
 
+// enrichConfigDefaults is called by the plugin when bootstrapping a job type's
+// default config from its descriptor. For admin_script, it fetches maintenance
+// scripts from the master and uses them as the script default.
+//
+// MIGRATION: This exists to help users migrate from master.toml [master.maintenance]
+// to the admin script plugin worker. Remove after March 2027.
+func (s *AdminServer) enrichConfigDefaults(cfg *plugin_pb.PersistedJobTypeConfig) *plugin_pb.PersistedJobTypeConfig {
+	if cfg.JobType != "admin_script" {
+		return cfg
+	}
+
+	var maintenanceScripts string
+	var sleepMinutes uint32
+	err := s.WithMasterClient(func(client master_pb.SeaweedClient) error {
+		ctx, cancel := context.WithTimeout(context.Background(), 10*time.Second)
+		defer cancel()
+		resp, err := client.GetMasterConfiguration(ctx, &master_pb.GetMasterConfigurationRequest{})
+		if err != nil {
+			return err
+		}
+		maintenanceScripts = resp.MaintenanceScripts
+		sleepMinutes = resp.MaintenanceSleepMinutes
+		return nil
+	})
+	if err != nil {
+		glog.V(1).Infof("Could not fetch master configuration for admin_script defaults: %v", err)
+		return cfg
+	}
+
+	script := cleanMaintenanceScript(maintenanceScripts)
+	if script == "" {
+		return cfg
+	}
+
+	interval := int64(sleepMinutes)
+	if interval <= 0 {
+		interval = clustermaintenance.DefaultMaintenanceSleepMinutes
+	}
+
+	glog.V(0).Infof("Enriching admin_script defaults from master maintenance scripts (interval=%dm)", interval)
+
+	if cfg.AdminConfigValues == nil {
+		cfg.AdminConfigValues = make(map[string]*plugin_pb.ConfigValue)
+	}
+	cfg.AdminConfigValues["script"] = &plugin_pb.ConfigValue{
+		Kind: &plugin_pb.ConfigValue_StringValue{StringValue: script},
+	}
+	cfg.AdminConfigValues["run_interval_minutes"] = &plugin_pb.ConfigValue{
+		Kind: &plugin_pb.ConfigValue_Int64Value{Int64Value: interval},
+	}
+	cfg.UpdatedBy = "master_migration"
+
+	return cfg
+}
+
+// cleanMaintenanceScript strips lock/unlock commands and normalizes a
+// maintenance script string for use with the admin script plugin worker.
+//
+// MIGRATION: Used by enrichConfigDefaults. Remove after March 2027.
+func cleanMaintenanceScript(script string) string {
+	script = strings.ReplaceAll(script, "\r\n", "\n")
+	var lines []string
+	for _, line := range strings.Split(script, "\n") {
+		trimmed := strings.TrimSpace(line)
+		if trimmed == "" || strings.HasPrefix(trimmed, "#") {
+			continue
+		}
+		// Strip inline comments (e.g., "lock # migration note")
+		if idx := strings.Index(trimmed, "#"); idx >= 0 {
+			trimmed = strings.TrimSpace(trimmed[:idx])
+			if trimmed == "" {
+				continue
+			}
+		}
+		firstToken := strings.ToLower(strings.Fields(trimmed)[0])
+		if firstToken == "lock" || firstToken == "unlock" {
+			continue
+		}
+		lines = append(lines, trimmed)
+	}
+	return strings.Join(lines, "\n")
+}
+
 // GetCredentialManager returns the credential manager
 func (s *AdminServer) GetCredentialManager() *credential.CredentialManager {
 	return s.credentialManager
@@ -284,8 +373,21 @@ func (s *AdminServer) GetCredentialManager() *credential.CredentialManager {
 
 // InvalidateCache method moved to cluster_topology.go
 
-// GetS3BucketsData retrieves all Object Store buckets and aggregates total storage metrics
-func (s *AdminServer) GetS3BucketsData() (S3BucketsData, error) {
+// GetS3BucketsData retrieves Object Store buckets with pagination and sorting
+func (s *AdminServer) GetS3BucketsData(page, pageSize int, sortBy, sortOrder string) (S3BucketsData, error) {
+	if page < 1 {
+		page = 1
+	}
+	if pageSize < 1 || pageSize > 1000 {
+		pageSize = 100
+	}
+	if sortBy == "" {
+		sortBy = "name"
+	}
+	if sortOrder == "" {
+		sortOrder = "asc"
+	}
+
 	buckets, err := s.GetS3Buckets()
 	if err != nil {
 		return S3BucketsData{}, err
@@ -296,14 +398,97 @@ func (s *AdminServer) GetS3BucketsData() (S3BucketsData, error) {
 		totalSize += bucket.PhysicalSize
 	}
 
+	totalBuckets := len(buckets)
+
+	// Sort buckets
+	s.sortBuckets(buckets, sortBy, sortOrder)
+
+	// Calculate pagination
+	totalPages := (totalBuckets + pageSize - 1) / pageSize
+	if totalPages == 0 {
+		totalPages = 1
+	}
+	if page > totalPages {
+		page = totalPages
+	}
+
+	startIndex := (page - 1) * pageSize
+	endIndex := startIndex + pageSize
+	if startIndex >= totalBuckets {
+		buckets = []S3Bucket{}
+	} else {
+		if endIndex > totalBuckets {
+			endIndex = totalBuckets
+		}
+		buckets = buckets[startIndex:endIndex]
+	}
+
 	return S3BucketsData{
 		Buckets:      buckets,
-		TotalBuckets: len(buckets),
+		TotalBuckets: totalBuckets,
 		TotalSize:    totalSize,
 		LastUpdated:  time.Now(),
+		CurrentPage:  page,
+		TotalPages:   totalPages,
+		PageSize:     pageSize,
+		SortBy:       sortBy,
+		SortOrder:    sortOrder,
 	}, nil
 }
 
+// sortBuckets sorts the bucket slice in place by the given field and order
+func (s *AdminServer) sortBuckets(buckets []S3Bucket, sortBy, sortOrder string) {
+	desc := sortOrder == "desc"
+	sort.Slice(buckets, func(i, j int) bool {
+		a, b := buckets[i], buckets[j]
+		switch sortBy {
+		case "owner":
+			if a.Owner != b.Owner {
+				if desc {
+					return a.Owner > b.Owner
+				}
+				return a.Owner < b.Owner
+			}
+		case "created":
+			if !a.CreatedAt.Equal(b.CreatedAt) {
+				if desc {
+					return a.CreatedAt.After(b.CreatedAt)
+				}
+				return a.CreatedAt.Before(b.CreatedAt)
+			}
+		case "objects":
+			if a.ObjectCount != b.ObjectCount {
+				if desc {
+					return a.ObjectCount > b.ObjectCount
+				}
+				return a.ObjectCount < b.ObjectCount
+			}
+		case "logical_size":
+			if a.LogicalSize != b.LogicalSize {
+				if desc {
+					return a.LogicalSize > b.LogicalSize
+				}
+				return a.LogicalSize < b.LogicalSize
+			}
+		case "physical_size":
+			if a.PhysicalSize != b.PhysicalSize {
+				if desc {
+					return a.PhysicalSize > b.PhysicalSize
+				}
+				return a.PhysicalSize < b.PhysicalSize
+			}
+		}
+		// Tie-breaker: sort by name (also the default/primary for sortBy=="name")
+		if a.Name != b.Name {
+			if desc {
+				return a.Name > b.Name
+			}
+			return a.Name < b.Name
+		}
+		return false
+	})
+}
+
 // GetS3Buckets retrieves all Object Store buckets from the filer and collects size/object data from collections
 func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 	var buckets []S3Bucket
@@ -319,28 +504,48 @@ func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 
 	// Now list buckets from the filer and match with collection data
 	err = s.WithFilerClient(func(client filer_pb.SeaweedFilerClient) error {
-		// List buckets by looking at the buckets directory
-		stream, err := client.ListEntries(context.Background(), &filer_pb.ListEntriesRequest{
-			Directory:          filerConfig.BucketsPath,
-			Prefix:             "",
-			StartFromFileName:  "",
-			InclusiveStartFrom: false,
-			Limit:              1000,
-		})
-		if err != nil {
-			return err
-		}
-
+		// Paginate through all buckets in the buckets directory
+		const listPageSize = 1000
+		startFrom := ""
+		var snapshotTsNs int64
 		for {
-			resp, err := stream.Recv()
+			stream, err := client.ListEntries(context.Background(), &filer_pb.ListEntriesRequest{
+				Directory:          filerConfig.BucketsPath,
+				Prefix:             "",
+				StartFromFileName:  startFrom,
+				InclusiveStartFrom: false,
+				Limit:              listPageSize,
+				SnapshotTsNs:       snapshotTsNs,
+			})
 			if err != nil {
-				if err.Error() == "EOF" {
-					break
-				}
 				return err
 			}
 
-			if resp.Entry.IsDirectory {
+			pageCount := 0
+			lastName := ""
+			for {
+				resp, err := stream.Recv()
+				if err != nil {
+					if errors.Is(err, io.EOF) {
+						break
+					}
+					return err
+				}
+
+				if snapshotTsNs == 0 && resp.SnapshotTsNs != 0 {
+					snapshotTsNs = resp.SnapshotTsNs
+				}
+
+				if resp.Entry == nil {
+					continue
+				}
+				lastName = resp.Entry.Name
+				pageCount++
+
+				if !resp.Entry.IsDirectory {
+					continue
+				}
+
 				bucketName := resp.Entry.Name
 				if strings.HasPrefix(bucketName, ".") {
 					// Skip internal/system directories from Object Store bucket listing.
@@ -393,13 +598,18 @@ func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 					}
 				}
 
+				var createdAt, lastModified time.Time
+				if resp.Entry.Attributes != nil {
+					createdAt = time.Unix(resp.Entry.Attributes.Crtime, 0)
+					lastModified = time.Unix(resp.Entry.Attributes.Mtime, 0)
+				}
 				bucket := S3Bucket{
 					Name:               bucketName,
-					CreatedAt:          time.Unix(resp.Entry.Attributes.Crtime, 0),
+					CreatedAt:          createdAt,
 					LogicalSize:        logicalSize,
 					PhysicalSize:       physicalSize,
 					ObjectCount:        objectCount,
-					LastModified:       time.Unix(resp.Entry.Attributes.Mtime, 0),
+					LastModified:       lastModified,
 					Quota:              quota,
 					QuotaEnabled:       quotaEnabled,
 					VersioningStatus:   versioningStatus,
@@ -410,6 +620,12 @@ func (s *AdminServer) GetS3Buckets() ([]S3Bucket, error) {
 				}
 				buckets = append(buckets, bucket)
 			}
+
+			// If we received fewer entries than the page size, we've listed everything
+			if pageCount < listPageSize {
+				break
+			}
+			startFrom = lastName
 		}
 
 		return nil
@@ -1286,6 +1502,9 @@ func (s *AdminServer) Shutdown() {
 
 	// Stop maintenance manager
 	s.StopMaintenanceManager()
+	if s.adminPresenceLock != nil {
+		s.adminPresenceLock.Stop()
+	}
 
 	if s.plugin != nil {
 		s.plugin.Shutdown()

weed/admin/dash/admin_server_seed_test.go

@@ -0,0 +1,77 @@
+// MIGRATION: Tests for enrichConfigDefaults helpers. Remove after March 2027.
+package dash
+
+import "testing"
+
+func TestCleanMaintenanceScript(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{
+			name:     "empty",
+			input:    "",
+			expected: "",
+		},
+		{
+			name:     "only lock unlock",
+			input:    "  lock\n  unlock\n",
+			expected: "",
+		},
+		{
+			name:     "strips lock and unlock",
+			input:    "  lock\n  ec.balance -apply\n  volume.fix.replication -apply\n  unlock\n",
+			expected: "ec.balance -apply\nvolume.fix.replication -apply",
+		},
+		{
+			name:     "case insensitive lock",
+			input:    "Lock\nec.balance -apply\nUNLOCK",
+			expected: "ec.balance -apply",
+		},
+		{
+			name:     "preserves comments removal",
+			input:    "lock\n# a comment\nec.balance -apply\nunlock",
+			expected: "ec.balance -apply",
+		},
+		{
+			name:     "no lock unlock present",
+			input:    "ec.balance -apply\nvolume.fix.replication -apply",
+			expected: "ec.balance -apply\nvolume.fix.replication -apply",
+		},
+		{
+			name:     "windows line endings",
+			input:    "lock\r\nec.balance -apply\r\nunlock\r\n",
+			expected: "ec.balance -apply",
+		},
+		{
+			name:     "lock with inline comment",
+			input:    "lock # migration\nec.balance -apply\nunlock # done",
+			expected: "ec.balance -apply",
+		},
+		{
+			name:     "command with inline comment preserved",
+			input:    "lock\nec.balance -apply # rebalance shards\nunlock",
+			expected: "ec.balance -apply",
+		},
+		{
+			name:     "only inline comment after stripping",
+			input:    "# full line comment\n  # indented comment\n",
+			expected: "",
+		},
+		{
+			name:     "typical master default",
+			input:    "\n  lock\n  ec.encode -fullPercent=95 -quietFor=1h\n  ec.rebuild -apply\n  ec.balance -apply\n  fs.log.purge -daysAgo=7\n  volume.deleteEmpty -quietFor=24h -apply\n  volume.balance -apply\n  volume.fix.replication -apply\n  s3.clean.uploads -timeAgo=24h\n  unlock\n",
+			expected: "ec.encode -fullPercent=95 -quietFor=1h\nec.rebuild -apply\nec.balance -apply\nfs.log.purge -daysAgo=7\nvolume.deleteEmpty -quietFor=24h -apply\nvolume.balance -apply\nvolume.fix.replication -apply\ns3.clean.uploads -timeAgo=24h",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := cleanMaintenanceScript(tt.input)
+			if got != tt.expected {
+				t.Errorf("cleanMaintenanceScript(%q) = %q, want %q", tt.input, got, tt.expected)
+			}
+		})
+	}
+}

weed/admin/dash/bucket_management.go

@@ -26,6 +26,15 @@ type S3BucketsData struct {
 	TotalBuckets int        `json:"total_buckets"`
 	TotalSize    int64      `json:"total_size"`
 	LastUpdated  time.Time  `json:"last_updated"`
+
+	// Pagination
+	CurrentPage int `json:"current_page"`
+	TotalPages  int `json:"total_pages"`
+	PageSize    int `json:"page_size"`
+
+	// Sorting
+	SortBy    string `json:"sort_by"`
+	SortOrder string `json:"sort_order"`
 }
 
 type CreateBucketRequest struct {
@@ -48,7 +57,7 @@ type CreateBucketRequest struct {
 func (s *AdminServer) ShowS3Buckets(w http.ResponseWriter, r *http.Request) {
 	username := UsernameFromContext(r.Context())
 
-	data, err := s.GetS3BucketsData()
+	data, err := s.GetS3BucketsData(1, 100, "name", "asc")
 	if err != nil {
 		writeJSONError(w, http.StatusInternalServerError, "Failed to get Object Store buckets: "+err.Error())
 		return

weed/admin/dash/group_management.go

@@ -0,0 +1,250 @@
+package dash
+
+import (
+	"context"
+	"errors"
+	"fmt"
+
+	"github.com/seaweedfs/seaweedfs/weed/credential"
+	"github.com/seaweedfs/seaweedfs/weed/glog"
+	"github.com/seaweedfs/seaweedfs/weed/pb/iam_pb"
+)
+
+// cloneGroup creates a deep copy of an iam_pb.Group to avoid mutating stored state.
+func cloneGroup(g *iam_pb.Group) *iam_pb.Group {
+	clone := &iam_pb.Group{
+		Name:     g.Name,
+		Disabled: g.Disabled,
+	}
+	if g.Members != nil {
+		clone.Members = make([]string, len(g.Members))
+		copy(clone.Members, g.Members)
+	}
+	if g.PolicyNames != nil {
+		clone.PolicyNames = make([]string, len(g.PolicyNames))
+		copy(clone.PolicyNames, g.PolicyNames)
+	}
+	return clone
+}
+
+func (s *AdminServer) GetGroups(ctx context.Context) ([]GroupData, error) {
+	if s.credentialManager == nil {
+		return nil, fmt.Errorf("credential manager not available")
+	}
+
+	groupNames, err := s.credentialManager.ListGroups(ctx)
+	if err != nil {
+		return nil, fmt.Errorf("failed to list groups: %w", err)
+	}
+
+	var groups []GroupData
+	for _, name := range groupNames {
+		g, err := s.credentialManager.GetGroup(ctx, name)
+		if err != nil {
+			if errors.Is(err, credential.ErrGroupNotFound) {
+				glog.V(1).Infof("Group %s listed but not found, skipping", name)
+				continue
+			}
+			return nil, fmt.Errorf("failed to get group %s: %w", name, err)
+		}
+		status := "enabled"
+		if g.Disabled {
+			status = "disabled"
+		}
+		groups = append(groups, GroupData{
+			Name:        g.Name,
+			MemberCount: len(g.Members),
+			PolicyCount: len(g.PolicyNames),
+			Status:      status,
+			Members:     g.Members,
+			PolicyNames: g.PolicyNames,
+		})
+	}
+	return groups, nil
+}
+
+func (s *AdminServer) GetGroupDetails(ctx context.Context, name string) (*GroupData, error) {
+	if s.credentialManager == nil {
+		return nil, fmt.Errorf("credential manager not available")
+	}
+
+	g, err := s.credentialManager.GetGroup(ctx, name)
+	if err != nil {
+		return nil, fmt.Errorf("failed to get group: %w", err)
+	}
+	status := "enabled"
+	if g.Disabled {
+		status = "disabled"
+	}
+	return &GroupData{
+		Name:        g.Name,
+		MemberCount: len(g.Members),
+		PolicyCount: len(g.PolicyNames),
+		Status:      status,
+		Members:     g.Members,
+		PolicyNames: g.PolicyNames,
+	}, nil
+}
+
+func (s *AdminServer) CreateGroup(ctx context.Context, name string) (*GroupData, error) {
+	if s.credentialManager == nil {
+		return nil, fmt.Errorf("credential manager not available")
+	}
+
+	group := &iam_pb.Group{Name: name}
+	if err := s.credentialManager.CreateGroup(ctx, group); err != nil {
+		return nil, fmt.Errorf("failed to create group: %w", err)
+	}
+	glog.V(1).Infof("Created group %s", group.Name)
+	return &GroupData{
+		Name:   group.Name,
+		Status: "enabled",
+	}, nil
+}
+
+func (s *AdminServer) DeleteGroup(ctx context.Context, name string) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	// Check for members and attached policies before deleting (same guards as IAM handlers)
+	g, err := s.credentialManager.GetGroup(ctx, name)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	if len(g.Members) > 0 {
+		return fmt.Errorf("cannot delete group %s: group has %d member(s): %w", name, len(g.Members), credential.ErrGroupNotEmpty)
+	}
+	if len(g.PolicyNames) > 0 {
+		return fmt.Errorf("cannot delete group %s: group has %d attached policy(ies): %w", name, len(g.PolicyNames), credential.ErrGroupNotEmpty)
+	}
+	if err := s.credentialManager.DeleteGroup(ctx, name); err != nil {
+		return fmt.Errorf("failed to delete group: %w", err)
+	}
+	glog.V(1).Infof("Deleted group %s", name)
+	return nil
+}
+
+func (s *AdminServer) AddGroupMember(ctx context.Context, groupName, username string) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	g, err := s.credentialManager.GetGroup(ctx, groupName)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	g = cloneGroup(g)
+	if _, err := s.credentialManager.GetUser(ctx, username); err != nil {
+		return fmt.Errorf("user %s not found: %w", username, err)
+	}
+	for _, m := range g.Members {
+		if m == username {
+			return nil // already a member
+		}
+	}
+	g.Members = append(g.Members, username)
+	if err := s.credentialManager.UpdateGroup(ctx, g); err != nil {
+		return fmt.Errorf("failed to update group: %w", err)
+	}
+	glog.V(1).Infof("Added user %s to group %s", username, groupName)
+	return nil
+}
+
+func (s *AdminServer) RemoveGroupMember(ctx context.Context, groupName, username string) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	g, err := s.credentialManager.GetGroup(ctx, groupName)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	g = cloneGroup(g)
+	found := false
+	var newMembers []string
+	for _, m := range g.Members {
+		if m == username {
+			found = true
+		} else {
+			newMembers = append(newMembers, m)
+		}
+	}
+	if !found {
+		return fmt.Errorf("user %s is not a member of group %s: %w", username, groupName, credential.ErrUserNotInGroup)
+	}
+	g.Members = newMembers
+	if err := s.credentialManager.UpdateGroup(ctx, g); err != nil {
+		return fmt.Errorf("failed to update group: %w", err)
+	}
+	glog.V(1).Infof("Removed user %s from group %s", username, groupName)
+	return nil
+}
+
+func (s *AdminServer) AttachGroupPolicy(ctx context.Context, groupName, policyName string) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	g, err := s.credentialManager.GetGroup(ctx, groupName)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	g = cloneGroup(g)
+	if _, err := s.credentialManager.GetPolicy(ctx, policyName); err != nil {
+		return fmt.Errorf("policy %s not found: %w", policyName, err)
+	}
+	for _, p := range g.PolicyNames {
+		if p == policyName {
+			return nil // already attached
+		}
+	}
+	g.PolicyNames = append(g.PolicyNames, policyName)
+	if err := s.credentialManager.UpdateGroup(ctx, g); err != nil {
+		return fmt.Errorf("failed to update group: %w", err)
+	}
+	glog.V(1).Infof("Attached policy %s to group %s", policyName, groupName)
+	return nil
+}
+
+func (s *AdminServer) DetachGroupPolicy(ctx context.Context, groupName, policyName string) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	g, err := s.credentialManager.GetGroup(ctx, groupName)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	g = cloneGroup(g)
+	found := false
+	var newPolicies []string
+	for _, p := range g.PolicyNames {
+		if p == policyName {
+			found = true
+		} else {
+			newPolicies = append(newPolicies, p)
+		}
+	}
+	if !found {
+		return fmt.Errorf("policy %s is not attached to group %s: %w", policyName, groupName, credential.ErrPolicyNotAttached)
+	}
+	g.PolicyNames = newPolicies
+	if err := s.credentialManager.UpdateGroup(ctx, g); err != nil {
+		return fmt.Errorf("failed to update group: %w", err)
+	}
+	glog.V(1).Infof("Detached policy %s from group %s", policyName, groupName)
+	return nil
+}
+
+func (s *AdminServer) SetGroupStatus(ctx context.Context, groupName string, enabled bool) error {
+	if s.credentialManager == nil {
+		return fmt.Errorf("credential manager not available")
+	}
+	g, err := s.credentialManager.GetGroup(ctx, groupName)
+	if err != nil {
+		return fmt.Errorf("failed to get group: %w", err)
+	}
+	g = cloneGroup(g)
+	g.Disabled = !enabled
+	if err := s.credentialManager.UpdateGroup(ctx, g); err != nil {
+		return fmt.Errorf("failed to update group: %w", err)
+	}
+	glog.V(1).Infof("Set group %s status to enabled=%v", groupName, enabled)
+	return nil
+}

weed/admin/dash/mq_management.go

@@ -324,7 +324,7 @@ func (s *AdminServer) GetConsumerGroupOffsets(namespace, topicName string) ([]Co
 			}
 
 			// Only process directories that are versions (start with "v")
-			if versionResp.Entry.IsDirectory && strings.HasPrefix(versionResp.Entry.Name, "v") {
+			if versionResp.Entry != nil && versionResp.Entry.IsDirectory && strings.HasPrefix(versionResp.Entry.Name, "v") {
 				versionDir := filepath.Join(topicDir, versionResp.Entry.Name)
 
 				// List all partition directories under the version directory (e.g., 0315-0630)
@@ -352,7 +352,7 @@ func (s *AdminServer) GetConsumerGroupOffsets(namespace, topicName string) ([]Co
 					}
 
 					// Only process directories that are partitions (format: NNNN-NNNN)
-					if partitionResp.Entry.IsDirectory {
+					if partitionResp.Entry != nil && partitionResp.Entry.IsDirectory {
 						// Parse partition range to get partition start ID (e.g., "0315-0630" -> 315)
 						var partitionStart, partitionStop int32
 						if n, err := fmt.Sscanf(partitionResp.Entry.Name, "%04d-%04d", &partitionStart, &partitionStop); n != 2 || err != nil {
@@ -387,11 +387,11 @@ func (s *AdminServer) GetConsumerGroupOffsets(namespace, topicName string) ([]Co
 							}
 
 							// Only process .offset files
-							if !offsetResp.Entry.IsDirectory && strings.HasSuffix(offsetResp.Entry.Name, ".offset") {
+							if offsetResp.Entry != nil && !offsetResp.Entry.IsDirectory && strings.HasSuffix(offsetResp.Entry.Name, ".offset") {
 								consumerGroup := strings.TrimSuffix(offsetResp.Entry.Name, ".offset")
 
 								// Read the offset value from the file
-								offsetData, err := filer.ReadInsideFiler(client, partitionDir, offsetResp.Entry.Name)
+								offsetData, err := filer.ReadInsideFiler(context.Background(), client, partitionDir, offsetResp.Entry.Name)
 								if err != nil {
 									glog.Warningf("Failed to read offset file %s: %v", offsetResp.Entry.Name, err)
 									continue
@@ -401,7 +401,10 @@ func (s *AdminServer) GetConsumerGroupOffsets(namespace, topicName string) ([]Co
 									offset := int64(util.BytesToUint64(offsetData))
 
 									// Get the file modification time
-									lastUpdated := time.Unix(offsetResp.Entry.Attributes.Mtime, 0)
+									var lastUpdated time.Time
+									if offsetResp.Entry.Attributes != nil {
+										lastUpdated = time.Unix(offsetResp.Entry.Attributes.Mtime, 0)
+									}
 
 									offsets = append(offsets, ConsumerGroupOffsetInfo{
 										ConsumerGroup: consumerGroup,

weed/admin/dash/plugin_api.go

@@ -867,6 +867,9 @@ func applyDescriptorDefaultsToPersistedConfig(
 		if runtime.PerWorkerExecutionConcurrency <= 0 {
 			runtime.PerWorkerExecutionConcurrency = defaults.PerWorkerExecutionConcurrency
 		}
+		if runtime.JobTypeMaxRuntimeSeconds <= 0 {
+			runtime.JobTypeMaxRuntimeSeconds = defaults.JobTypeMaxRuntimeSeconds
+		}
 		if runtime.RetryBackoffSeconds <= 0 {
 			runtime.RetryBackoffSeconds = defaults.RetryBackoffSeconds
 		}

weed/admin/dash/topic_retention.go

@@ -151,17 +151,21 @@ func (p *TopicRetentionPurger) purgeTopicData(topicRetention TopicRetentionConfi
 			}
 
 			// Only process directories that are versions (start with "v")
-			if versionResp.Entry.IsDirectory && strings.HasPrefix(versionResp.Entry.Name, "v") {
+			if versionResp.Entry != nil && versionResp.Entry.IsDirectory && strings.HasPrefix(versionResp.Entry.Name, "v") {
 				versionTime, err := p.parseVersionTime(versionResp.Entry.Name)
 				if err != nil {
 					glog.Warningf("Failed to parse version time from %s: %v", versionResp.Entry.Name, err)
 					continue
 				}
 
+				var modTime time.Time
+				if versionResp.Entry.Attributes != nil {
+					modTime = time.Unix(versionResp.Entry.Attributes.Mtime, 0)
+				}
 				versionDirs = append(versionDirs, VersionDirInfo{
 					Name:        versionResp.Entry.Name,
 					VersionTime: versionTime,
-					ModTime:     time.Unix(versionResp.Entry.Attributes.Mtime, 0),
+					ModTime:     modTime,
 				})
 			}
 		}
@@ -260,6 +264,9 @@ func (p *TopicRetentionPurger) deleteDirectoryRecursively(client filer_pb.Seawee
 			return fmt.Errorf("failed to receive entries: %w", err)
 		}
 
+		if resp.Entry == nil {
+			continue
+		}
 		entryPath := filepath.Join(dirPath, resp.Entry.Name)
 
 		if resp.Entry.IsDirectory {

weed/admin/dash/types.go

@@ -589,6 +589,30 @@ type UpdateServiceAccountRequest struct {
 	Expiration  string `json:"expiration,omitempty"`
 }
 
+// Group management structures
+type GroupData struct {
+	Name        string   `json:"name"`
+	MemberCount int      `json:"member_count"`
+	PolicyCount int      `json:"policy_count"`
+	Status      string   `json:"status"` // "enabled" or "disabled"
+	Members     []string `json:"members"`
+	PolicyNames []string `json:"policy_names"`
+}
+
+type GroupsPageData struct {
+	Username          string      `json:"username"`
+	Groups            []GroupData `json:"groups"`
+	TotalGroups       int         `json:"total_groups"`
+	ActiveGroups      int         `json:"active_groups"`
+	AvailableUsers    []string    `json:"available_users"`
+	AvailablePolicies []string    `json:"available_policies"`
+	LastUpdated       time.Time   `json:"last_updated"`
+}
+
+type CreateGroupRequest struct {
+	Name string `json:"name"`
+}
+
 // STS Configuration display types
 type STSConfigData struct {
 	Enabled       bool      `json:"enabled"`

weed/admin/dash/user_management.go

@@ -4,13 +4,21 @@ import (
 	"context"
 	"crypto/rand"
 	"encoding/base64"
+	"errors"
 	"fmt"
+	"strings"
 	"time"
 
 	"github.com/seaweedfs/seaweedfs/weed/credential"
 	"github.com/seaweedfs/seaweedfs/weed/pb/iam_pb"
 )
 
+var (
+	ErrAccessKeyInUse = errors.New("access key already in use")
+	ErrUserNotFound   = errors.New("user not found")
+	ErrInvalidInput   = errors.New("invalid input")
+)
+
 // CreateObjectStoreUser creates a new user using the credential manager
 func (s *AdminServer) CreateObjectStoreUser(req CreateUserRequest) (*ObjectStoreUser, error) {
 	if s.credentialManager == nil {
@@ -187,6 +195,24 @@ func (s *AdminServer) GetObjectStoreUserDetails(username string) (*UserDetails,
 		details.Email = identity.Account.EmailAddress
 	}
 
+	// Look up groups the user belongs to
+	groupNames, err := s.credentialManager.ListGroups(ctx)
+	if err != nil {
+		return nil, fmt.Errorf("failed to list groups: %w", err)
+	}
+	for _, gName := range groupNames {
+		g, err := s.credentialManager.GetGroup(ctx, gName)
+		if err != nil {
+			return nil, fmt.Errorf("failed to get group %s: %w", gName, err)
+		}
+		for _, member := range g.Members {
+			if member == username {
+				details.Groups = append(details.Groups, gName)
+				break
+			}
+		}
+	}
+
 	// Convert credentials to access key info
 	for _, cred := range identity.Credentials {
 		details.AccessKeys = append(details.AccessKeys, AccessKeyInfo{
@@ -201,7 +227,7 @@ func (s *AdminServer) GetObjectStoreUserDetails(username string) (*UserDetails,
 }
 
 // CreateAccessKey creates a new access key for a user
-func (s *AdminServer) CreateAccessKey(username string) (*AccessKeyInfo, error) {
+func (s *AdminServer) CreateAccessKey(username string, req *CreateAccessKeyRequest) (*AccessKeyInfo, error) {
 	if s.credentialManager == nil {
 		return nil, fmt.Errorf("credential manager not available")
 	}
@@ -212,14 +238,41 @@ func (s *AdminServer) CreateAccessKey(username string) (*AccessKeyInfo, error) {
 	_, err := s.credentialManager.GetUser(ctx, username)
 	if err != nil {
 		if err == credential.ErrUserNotFound {
-			return nil, fmt.Errorf("user %s not found", username)
+			return nil, fmt.Errorf("user %s: %w", username, ErrUserNotFound)
 		}
 		return nil, fmt.Errorf("failed to get user: %w", err)
 	}
 
-	// Generate new access key
-	accessKey := generateAccessKey()
-	secretKey := generateSecretKey()
+	if req == nil {
+		req = &CreateAccessKeyRequest{}
+	}
+
+	// Validate provided keys
+	if req.AccessKey != "" && (len(req.AccessKey) < 4 || len(req.AccessKey) > 128) {
+		return nil, fmt.Errorf("access key must be between 4 and 128 characters: %w", ErrInvalidInput)
+	}
+	if req.SecretKey != "" && (len(req.SecretKey) < 8 || len(req.SecretKey) > 128) {
+		return nil, fmt.Errorf("secret key must be between 8 and 128 characters: %w", ErrInvalidInput)
+	}
+
+	// Use provided keys or generate new ones
+	accessKey := req.AccessKey
+	if accessKey == "" {
+		accessKey = generateAccessKey()
+	}
+	secretKey := req.SecretKey
+	if secretKey == "" {
+		secretKey = generateSecretKey()
+	}
+
+	// Verify access key is globally unique
+	existingUser, err := s.credentialManager.GetUserByAccessKey(ctx, accessKey)
+	if existingUser != nil {
+		return nil, ErrAccessKeyInUse
+	}
+	if err != nil && !errors.Is(err, credential.ErrAccessKeyNotFound) && !isNotFoundError(err) {
+		return nil, fmt.Errorf("failed to check access key uniqueness: %w", err)
+	}
 
 	credential := &iam_pb.Credential{
 		AccessKey: accessKey,
@@ -364,6 +417,12 @@ func (s *AdminServer) UpdateUserPolicies(username string, actions []string) erro
 	return nil
 }
 
+// isNotFoundError checks for "not found" in the error message as a fallback
+// for stores (e.g. gRPC) that don't return the credential.ErrAccessKeyNotFound sentinel.
+func isNotFoundError(err error) bool {
+	return err != nil && strings.Contains(strings.ToLower(err.Error()), "not found")
+}
+
 // Helper functions for generating keys and IDs
 func generateAccessKey() string {
 	// Generate 20-character access key (AWS standard)

weed/admin/dash/volume_management.go

@@ -457,6 +457,7 @@ func (s *AdminServer) GetClusterVolumeServers() (*ClusterVolumeServersData, erro
 
 						// Process disk information
 						for _, diskInfo := range node.DiskInfos {
+							vs.MaxVolumes += int(diskInfo.MaxVolumeCount)
 							vs.DiskCapacity += int64(diskInfo.MaxVolumeCount) * int64(volumeSizeLimitMB) * 1024 * 1024 // Use actual volume size limit
 
 							// Count regular volumes and calculate disk usage

weed/admin/handlers/admin_handlers.go

@@ -3,6 +3,7 @@ package handlers
 import (
 	"net/http"
 	"net/url"
+	"strconv"
 	"time"
 
 	"github.com/gorilla/mux"
@@ -28,6 +29,7 @@ type AdminHandlers struct {
 	pluginHandlers         *PluginHandlers
 	mqHandlers             *MessageQueueHandlers
 	serviceAccountHandlers *ServiceAccountHandlers
+	groupHandlers          *GroupHandlers
 }
 
 // NewAdminHandlers creates a new instance of AdminHandlers
@@ -40,6 +42,7 @@ func NewAdminHandlers(adminServer *dash.AdminServer, store sessions.Store) *Admi
 	pluginHandlers := NewPluginHandlers(adminServer)
 	mqHandlers := NewMessageQueueHandlers(adminServer)
 	serviceAccountHandlers := NewServiceAccountHandlers(adminServer)
+	groupHandlers := NewGroupHandlers(adminServer)
 	return &AdminHandlers{
 		adminServer:            adminServer,
 		sessionStore:           store,
@@ -51,6 +54,7 @@ func NewAdminHandlers(adminServer *dash.AdminServer, store sessions.Store) *Admi
 		pluginHandlers:         pluginHandlers,
 		mqHandlers:             mqHandlers,
 		serviceAccountHandlers: serviceAccountHandlers,
+		groupHandlers:          groupHandlers,
 	}
 }
 
@@ -104,6 +108,7 @@ func (h *AdminHandlers) registerUIRoutes(r *mux.Router) {
 	r.HandleFunc("/object-store/buckets/{bucket}", h.ShowBucketDetails).Methods(http.MethodGet)
 	r.HandleFunc("/object-store/users", h.userHandlers.ShowObjectStoreUsers).Methods(http.MethodGet)
 	r.HandleFunc("/object-store/policies", h.policyHandlers.ShowPolicies).Methods(http.MethodGet)
+	r.HandleFunc("/object-store/groups", h.groupHandlers.ShowGroups).Methods(http.MethodGet)
 	r.HandleFunc("/object-store/service-accounts", h.serviceAccountHandlers.ShowServiceAccounts).Methods(http.MethodGet)
 	r.HandleFunc("/object-store/s3tables/buckets", h.ShowS3TablesBuckets).Methods(http.MethodGet)
 	r.HandleFunc("/object-store/s3tables/buckets/{bucket}/namespaces", h.ShowS3TablesNamespaces).Methods(http.MethodGet)
@@ -185,6 +190,19 @@ func (h *AdminHandlers) registerAPIRoutes(api *mux.Router, enforceWrite bool) {
 	saApi.Handle("/{id}", wrapWrite(h.serviceAccountHandlers.UpdateServiceAccount)).Methods(http.MethodPut)
 	saApi.Handle("/{id}", wrapWrite(h.serviceAccountHandlers.DeleteServiceAccount)).Methods(http.MethodDelete)
 
+	groupsApi := api.PathPrefix("/groups").Subrouter()
+	groupsApi.HandleFunc("", h.groupHandlers.GetGroups).Methods(http.MethodGet)
+	groupsApi.Handle("", wrapWrite(h.groupHandlers.CreateGroup)).Methods(http.MethodPost)
+	groupsApi.HandleFunc("/{name}", h.groupHandlers.GetGroupDetails).Methods(http.MethodGet)
+	groupsApi.Handle("/{name}", wrapWrite(h.groupHandlers.DeleteGroup)).Methods(http.MethodDelete)
+	groupsApi.Handle("/{name}/status", wrapWrite(h.groupHandlers.SetGroupStatus)).Methods(http.MethodPut)
+	groupsApi.HandleFunc("/{name}/members", h.groupHandlers.GetGroupMembers).Methods(http.MethodGet)
+	groupsApi.Handle("/{name}/members", wrapWrite(h.groupHandlers.AddGroupMember)).Methods(http.MethodPost)
+	groupsApi.Handle("/{name}/members/{username}", wrapWrite(h.groupHandlers.RemoveGroupMember)).Methods(http.MethodDelete)
+	groupsApi.HandleFunc("/{name}/policies", h.groupHandlers.GetGroupPolicies).Methods(http.MethodGet)
+	groupsApi.Handle("/{name}/policies", wrapWrite(h.groupHandlers.AttachGroupPolicy)).Methods(http.MethodPost)
+	groupsApi.Handle("/{name}/policies/{policyName}", wrapWrite(h.groupHandlers.DetachGroupPolicy)).Methods(http.MethodDelete)
+
 	policyApi := api.PathPrefix("/object-store/policies").Subrouter()
 	policyApi.HandleFunc("", h.policyHandlers.GetPolicies).Methods(http.MethodGet)
 	policyApi.Handle("", wrapWrite(h.policyHandlers.CreatePolicy)).Methods(http.MethodPost)
@@ -276,8 +294,26 @@ func (h *AdminHandlers) ShowDashboard(w http.ResponseWriter, r *http.Request) {
 
 // ShowS3Buckets renders the Object Store buckets management page
 func (h *AdminHandlers) ShowS3Buckets(w http.ResponseWriter, r *http.Request) {
-	// Get Object Store buckets data from the server
-	s3Data := h.getS3BucketsData(r)
+	// Get pagination and sorting parameters from query string
+	page := 1
+	if p := r.URL.Query().Get("page"); p != "" {
+		if parsed, err := strconv.Atoi(p); err == nil && parsed > 0 {
+			page = parsed
+		}
+	}
+
+	pageSize := 100
+	if ps := r.URL.Query().Get("pageSize"); ps != "" {
+		if parsed, err := strconv.Atoi(ps); err == nil && parsed > 0 && parsed <= 1000 {
+			pageSize = parsed
+		}
+	}
+
+	sortBy := defaultQuery(r.URL.Query().Get("sortBy"), "name")
+	sortOrder := defaultQuery(r.URL.Query().Get("sortOrder"), "asc")
+
+	// Get Object Store buckets data with pagination
+	s3Data := h.getS3BucketsData(r, page, pageSize, sortBy, sortOrder)
 	username := h.getUsername(r)
 
 	// Render HTML template
@@ -444,15 +480,15 @@ func (h *AdminHandlers) ShowBucketDetails(w http.ResponseWriter, r *http.Request
 	writeJSON(w, http.StatusOK, details)
 }
 
-// getS3BucketsData retrieves Object Store buckets data from the server
-func (h *AdminHandlers) getS3BucketsData(r *http.Request) dash.S3BucketsData {
+// getS3BucketsData retrieves Object Store buckets data from the server with pagination
+func (h *AdminHandlers) getS3BucketsData(r *http.Request, page, pageSize int, sortBy, sortOrder string) dash.S3BucketsData {
 	username := dash.UsernameFromContext(r.Context())
 	if username == "" {
 		username = "admin"
 	}
 
 	// Get Object Store buckets data
-	data, err := h.adminServer.GetS3BucketsData()
+	data, err := h.adminServer.GetS3BucketsData(page, pageSize, sortBy, sortOrder)
 	if err != nil {
 		// Return empty data on error
 		return dash.S3BucketsData{
@@ -461,6 +497,11 @@ func (h *AdminHandlers) getS3BucketsData(r *http.Request) dash.S3BucketsData {
 			TotalBuckets: 0,
 			TotalSize:    0,
 			LastUpdated:  time.Now(),
+			CurrentPage:  1,
+			TotalPages:   1,
+			PageSize:     pageSize,
+			SortBy:       sortBy,
+			SortOrder:    sortOrder,
 		}
 	}
 

weed/admin/handlers/group_handlers.go

@@ -0,0 +1,271 @@
+package handlers
+
+import (
+	"bytes"
+	"errors"
+	"net/http"
+	"time"
+
+	"github.com/gorilla/mux"
+	"github.com/seaweedfs/seaweedfs/weed/admin/dash"
+	"github.com/seaweedfs/seaweedfs/weed/admin/view/app"
+	"github.com/seaweedfs/seaweedfs/weed/admin/view/layout"
+	"github.com/seaweedfs/seaweedfs/weed/credential"
+	"github.com/seaweedfs/seaweedfs/weed/glog"
+)
+
+func groupErrorToHTTPStatus(err error) int {
+	if errors.Is(err, credential.ErrGroupNotFound) {
+		return http.StatusNotFound
+	}
+	if errors.Is(err, credential.ErrGroupAlreadyExists) {
+		return http.StatusConflict
+	}
+	if errors.Is(err, credential.ErrUserNotInGroup) {
+		return http.StatusBadRequest
+	}
+	if errors.Is(err, credential.ErrPolicyNotAttached) {
+		return http.StatusBadRequest
+	}
+	if errors.Is(err, credential.ErrUserNotFound) {
+		return http.StatusNotFound
+	}
+	if errors.Is(err, credential.ErrPolicyNotFound) {
+		return http.StatusNotFound
+	}
+	if errors.Is(err, credential.ErrGroupNotEmpty) {
+		return http.StatusConflict
+	}
+	return http.StatusInternalServerError
+}
+
+type GroupHandlers struct {
+	adminServer *dash.AdminServer
+}
+
+func NewGroupHandlers(adminServer *dash.AdminServer) *GroupHandlers {
+	return &GroupHandlers{adminServer: adminServer}
+}
+
+func (h *GroupHandlers) ShowGroups(w http.ResponseWriter, r *http.Request) {
+	data, err := h.getGroupsPageData(r)
+	if err != nil {
+		glog.Errorf("Failed to get groups data: %v", err)
+		writeJSONError(w, http.StatusInternalServerError, "Failed to load groups: "+err.Error())
+		return
+	}
+
+	var buf bytes.Buffer
+	component := app.Groups(data)
+	viewCtx := layout.NewViewContext(r, dash.UsernameFromContext(r.Context()), dash.CSRFTokenFromContext(r.Context()))
+	layoutComponent := layout.Layout(viewCtx, component)
+	if err := layoutComponent.Render(r.Context(), &buf); err != nil {
+		glog.Errorf("Failed to render groups template: %v", err)
+		w.WriteHeader(http.StatusInternalServerError)
+		return
+	}
+	w.Header().Set("Content-Type", "text/html")
+	_, _ = w.Write(buf.Bytes())
+}
+
+func (h *GroupHandlers) GetGroups(w http.ResponseWriter, r *http.Request) {
+	groups, err := h.adminServer.GetGroups(r.Context())
+	if err != nil {
+		glog.Errorf("Failed to get groups: %v", err)
+		writeJSONError(w, http.StatusInternalServerError, "Failed to get groups")
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]interface{}{"groups": groups})
+}
+
+func (h *GroupHandlers) CreateGroup(w http.ResponseWriter, r *http.Request) {
+	var req dash.CreateGroupRequest
+	if err := decodeJSONBody(newJSONMaxReader(w, r), &req); err != nil {
+		writeJSONError(w, http.StatusBadRequest, "Invalid request: "+err.Error())
+		return
+	}
+	if req.Name == "" {
+		writeJSONError(w, http.StatusBadRequest, "Group name is required")
+		return
+	}
+	group, err := h.adminServer.CreateGroup(r.Context(), req.Name)
+	if err != nil {
+		glog.Errorf("Failed to create group: %v", err)
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to create group: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, group)
+}
+
+func (h *GroupHandlers) GetGroupDetails(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	group, err := h.adminServer.GetGroupDetails(r.Context(), name)
+	if err != nil {
+		glog.Errorf("Failed to get group details: %v", err)
+		status := groupErrorToHTTPStatus(err)
+		msg := "Failed to retrieve group"
+		if status == http.StatusNotFound {
+			msg = "Group not found"
+		}
+		writeJSONError(w, status, msg)
+		return
+	}
+	writeJSON(w, http.StatusOK, group)
+}
+
+func (h *GroupHandlers) DeleteGroup(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	if err := h.adminServer.DeleteGroup(r.Context(), name); err != nil {
+		glog.Errorf("Failed to delete group: %v", err)
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to delete group: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Group deleted successfully"})
+}
+
+func (h *GroupHandlers) GetGroupMembers(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	group, err := h.adminServer.GetGroupDetails(r.Context(), name)
+	if err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to get group: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]interface{}{"members": group.Members})
+}
+
+func (h *GroupHandlers) AddGroupMember(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	var req struct {
+		Username string `json:"username"`
+	}
+	if err := decodeJSONBody(newJSONMaxReader(w, r), &req); err != nil {
+		writeJSONError(w, http.StatusBadRequest, "Invalid request: "+err.Error())
+		return
+	}
+	if req.Username == "" {
+		writeJSONError(w, http.StatusBadRequest, "Username is required")
+		return
+	}
+	if err := h.adminServer.AddGroupMember(r.Context(), name, req.Username); err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to add member: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Member added successfully"})
+}
+
+func (h *GroupHandlers) RemoveGroupMember(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	username := mux.Vars(r)["username"]
+	if err := h.adminServer.RemoveGroupMember(r.Context(), name, username); err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to remove member: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Member removed successfully"})
+}
+
+func (h *GroupHandlers) GetGroupPolicies(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	group, err := h.adminServer.GetGroupDetails(r.Context(), name)
+	if err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to get group: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]interface{}{"policies": group.PolicyNames})
+}
+
+func (h *GroupHandlers) AttachGroupPolicy(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	var req struct {
+		PolicyName string `json:"policy_name"`
+	}
+	if err := decodeJSONBody(newJSONMaxReader(w, r), &req); err != nil {
+		writeJSONError(w, http.StatusBadRequest, "Invalid request: "+err.Error())
+		return
+	}
+	if req.PolicyName == "" {
+		writeJSONError(w, http.StatusBadRequest, "Policy name is required")
+		return
+	}
+	if err := h.adminServer.AttachGroupPolicy(r.Context(), name, req.PolicyName); err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to attach policy: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Policy attached successfully"})
+}
+
+func (h *GroupHandlers) DetachGroupPolicy(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	policyName := mux.Vars(r)["policyName"]
+	if err := h.adminServer.DetachGroupPolicy(r.Context(), name, policyName); err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to detach policy: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Policy detached successfully"})
+}
+
+func (h *GroupHandlers) SetGroupStatus(w http.ResponseWriter, r *http.Request) {
+	name := mux.Vars(r)["name"]
+	var req struct {
+		Enabled *bool `json:"enabled"`
+	}
+	if err := decodeJSONBody(newJSONMaxReader(w, r), &req); err != nil {
+		writeJSONError(w, http.StatusBadRequest, "Invalid request: "+err.Error())
+		return
+	}
+	if req.Enabled == nil {
+		writeJSONError(w, http.StatusBadRequest, "enabled field is required")
+		return
+	}
+	if err := h.adminServer.SetGroupStatus(r.Context(), name, *req.Enabled); err != nil {
+		writeJSONError(w, groupErrorToHTTPStatus(err), "Failed to update group status: "+err.Error())
+		return
+	}
+	writeJSON(w, http.StatusOK, map[string]string{"message": "Group status updated"})
+}
+
+func (h *GroupHandlers) getGroupsPageData(r *http.Request) (dash.GroupsPageData, error) {
+	username := dash.UsernameFromContext(r.Context())
+	if username == "" {
+		username = "admin"
+	}
+
+	groups, err := h.adminServer.GetGroups(r.Context())
+	if err != nil {
+		return dash.GroupsPageData{}, err
+	}
+
+	activeCount := 0
+	for _, g := range groups {
+		if g.Status == "enabled" {
+			activeCount++
+		}
+	}
+
+	// Get available users for dropdown
+	var availableUsers []string
+	users, err := h.adminServer.GetObjectStoreUsers(r.Context())
+	if err == nil {
+		for _, user := range users {
+			availableUsers = append(availableUsers, user.Username)
+		}
+	}
+
+	// Get available policies for dropdown
+	var availablePolicies []string
+	policies, err := h.adminServer.GetPolicies()
+	if err == nil {
+		for _, p := range policies {
+			availablePolicies = append(availablePolicies, p.Name)
+		}
+	}
+
+	return dash.GroupsPageData{
+		Username:          username,
+		Groups:            groups,
+		TotalGroups:       len(groups),
+		ActiveGroups:      activeCount,
+		AvailableUsers:    availableUsers,
+		AvailablePolicies: availablePolicies,
+		LastUpdated:       time.Now(),
+	}, nil
+}

weed/admin/handlers/user_handlers.go

@@ -1,7 +1,9 @@
 package handlers
 
 import (
+	"errors"
 	"fmt"
+	"io"
 	"net/http"
 	"time"
 
@@ -155,10 +157,30 @@ func (h *UserHandlers) CreateAccessKey(w http.ResponseWriter, r *http.Request) {
 		return
 	}
 
-	accessKey, err := h.adminServer.CreateAccessKey(username)
+	var req *dash.CreateAccessKeyRequest
+	var body dash.CreateAccessKeyRequest
+	if err := decodeJSONBody(newJSONMaxReader(w, r), &body); err != nil {
+		if !errors.Is(err, io.EOF) {
+			writeJSONError(w, http.StatusBadRequest, "Invalid request: "+err.Error())
+			return
+		}
+		// Empty body: auto-generate both keys
+	} else {
+		req = &body
+	}
+
+	accessKey, err := h.adminServer.CreateAccessKey(username, req)
 	if err != nil {
 		glog.Errorf("Failed to create access key for user %s: %v", username, err)
-		writeJSONError(w, http.StatusInternalServerError, "Failed to create access key: "+err.Error())
+		if errors.Is(err, dash.ErrAccessKeyInUse) {
+			writeJSONError(w, http.StatusConflict, err.Error())
+		} else if errors.Is(err, dash.ErrUserNotFound) {
+			writeJSONError(w, http.StatusNotFound, err.Error())
+		} else if errors.Is(err, dash.ErrInvalidInput) {
+			writeJSONError(w, http.StatusBadRequest, err.Error())
+		} else {
+			writeJSONError(w, http.StatusInternalServerError, "Failed to create access key: "+err.Error())
+		}
 		return
 	}
 

weed/admin/maintenance/maintenance_integration.go

@@ -5,6 +5,7 @@ import (
 
 	"github.com/seaweedfs/seaweedfs/weed/admin/topology"
 	"github.com/seaweedfs/seaweedfs/weed/glog"
+	"github.com/seaweedfs/seaweedfs/weed/pb"
 	"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
 	"github.com/seaweedfs/seaweedfs/weed/worker/tasks"
 	"github.com/seaweedfs/seaweedfs/weed/worker/types"
@@ -229,6 +230,12 @@ func (s *MaintenanceIntegration) ScanWithTaskDetectors(volumeMetrics []*types.Vo
 			continue
 		}
 
+		// Cancel stale pending tasks for this type before re-detection
+		maintenanceType := s.taskTypeMap[taskType]
+		if cancelled := s.maintenanceQueue.CancelPendingTasksByType(maintenanceType); cancelled > 0 {
+			glog.Infof("Cancelled %d stale pending %s tasks before re-detection", cancelled, taskType)
+		}
+
 		glog.V(2).Infof("Running detection for task type: %s", taskType)
 
 		results, err := detector.ScanForTasks(filteredMetrics, clusterInfo)
@@ -528,10 +535,15 @@ func (s *MaintenanceIntegration) SyncTask(task *MaintenanceTask) {
 		// Volume size is not currently used for Balance/Vacuum impact and is not stored in MaintenanceTask
 		sourceImpact, targetImpact := topology.CalculateTaskStorageImpact(topology.TaskType(string(taskType)), 0)
 
-		// Use unified sources and targets from TaskParams
+		// Use unified sources and targets from TaskParams.
+		// Task protos store ServerAddresses (with gRPC port, e.g., "host:port.grpcPort")
+		// but the topology indexes disks by NodeId (e.g., "host:port").
+		// Strip the gRPC port suffix via ToHttpAddress() to match the topology key.
 		for _, src := range task.TypedParams.Sources {
+			resolvedSrc := pb.ServerAddress(src.Node).ToHttpAddress()
+			glog.V(2).Infof("SyncTask %s: source proto Node=%q resolved to %q, diskId=%d", task.ID, src.Node, resolvedSrc, src.DiskId)
 			sources = append(sources, topology.TaskSource{
-				SourceServer:  src.Node,
+				SourceServer:  resolvedSrc,
 				SourceDisk:    src.DiskId,
 				StorageChange: sourceImpact,
 			})
@@ -539,8 +551,10 @@ func (s *MaintenanceIntegration) SyncTask(task *MaintenanceTask) {
 			estimatedSize += int64(src.EstimatedSize)
 		}
 		for _, target := range task.TypedParams.Targets {
+			resolvedTarget := pb.ServerAddress(target.Node).ToHttpAddress()
+			glog.V(2).Infof("SyncTask %s: target proto Node=%q resolved to %q, diskId=%d", task.ID, target.Node, resolvedTarget, target.DiskId)
 			destinations = append(destinations, topology.TaskDestination{
-				TargetServer:  target.Node,
+				TargetServer:  resolvedTarget,
 				TargetDisk:    target.DiskId,
 				StorageChange: targetImpact,
 			})

weed/admin/maintenance/maintenance_manager.go

@@ -297,7 +297,7 @@ func (mm *MaintenanceManager) logTopologyStatus() {
 	errorCount := mm.errorCount
 	mm.mutex.RUnlock()
 
-	glog.V(0).Infof("Topology status: %d nodes, %d disks, %d workers, %d pending tasks, %d running tasks, errors: %d",
+	glog.V(1).Infof("Topology status: %d nodes, %d disks, %d workers, %d pending tasks, %d running tasks, errors: %d",
 		nodeCount, diskCount, workerCount,
 		stats.TasksByStatus[TaskStatusPending],
 		stats.TasksByStatus[TaskStatusInProgress]+stats.TasksByStatus[TaskStatusAssigned],

weed/admin/maintenance/maintenance_queue.go

@@ -4,6 +4,7 @@ import (
 	"crypto/rand"
 	"fmt"
 	"sort"
+	"strings"
 	"time"
 
 	"github.com/seaweedfs/seaweedfs/weed/glog"
@@ -230,6 +231,46 @@ func (mq *MaintenanceQueue) hasDuplicateTask(newTask *MaintenanceTask) bool {
 	return false
 }
 
+// CancelPendingTasksByType cancels all pending tasks of a given type.
+// This is called before each detection cycle to ensure stale proposals
+// from previous cycles are cleaned up before creating new ones.
+func (mq *MaintenanceQueue) CancelPendingTasksByType(taskType MaintenanceTaskType) int {
+	mq.mutex.Lock()
+
+	var remaining []*MaintenanceTask
+	var cancelledSnapshots []*MaintenanceTask
+	cancelled := 0
+	for _, task := range mq.pendingTasks {
+		if task.Type == taskType {
+			task.Status = TaskStatusCancelled
+			now := time.Now()
+			task.CompletedAt = &now
+			cancelled++
+			cancelledSnapshots = append(cancelledSnapshots, snapshotTask(task))
+			glog.V(1).Infof("Cancelled stale pending task %s (%s) for volume %d before re-detection",
+				task.ID, task.Type, task.VolumeID)
+
+			// Release capacity in ActiveTopology and remove pending operation
+			if mq.integration != nil {
+				if at := mq.integration.GetActiveTopology(); at != nil {
+					_ = at.CompleteTask(task.ID)
+				}
+			}
+			mq.removePendingOperation(task.ID)
+		} else {
+			remaining = append(remaining, task)
+		}
+	}
+	mq.pendingTasks = remaining
+	mq.mutex.Unlock()
+
+	// Persist cancelled state outside the lock to avoid blocking
+	for _, snapshot := range cancelledSnapshots {
+		mq.saveTaskState(snapshot)
+	}
+	return cancelled
+}
+
 // AddTasksFromResults converts detection results to tasks and adds them to the queue
 func (mq *MaintenanceQueue) AddTasksFromResults(results []*TaskDetectionResult) {
 	for _, result := range results {
@@ -455,8 +496,8 @@ func (mq *MaintenanceQueue) CompleteTask(taskID string, error string) {
 		task.Status = TaskStatusFailed
 		task.Error = error
 
-		// Check if task should be retried
-		if task.RetryCount < task.MaxRetries {
+		// Check if task should be retried (skip retry for permanent errors)
+		if task.RetryCount < task.MaxRetries && !isNonRetriableError(error) {
 			// Record unassignment due to failure/retry
 			if task.WorkerID != "" && len(task.AssignmentHistory) > 0 {
 				lastAssignment := task.AssignmentHistory[len(task.AssignmentHistory)-1]
@@ -559,6 +600,12 @@ func (mq *MaintenanceQueue) CompleteTask(taskID string, error string) {
 	}
 }
 
+// isNonRetriableError returns true for errors that will never succeed on retry,
+// such as when the volume doesn't exist on the source server.
+func isNonRetriableError(errMsg string) bool {
+	return strings.Contains(errMsg, "not found")
+}
+
 // UpdateTaskProgress updates the progress of a running task
 func (mq *MaintenanceQueue) UpdateTaskProgress(taskID string, progress float64) {
 	mq.mutex.Lock()

weed/admin/plugin/config_store.go

@@ -200,6 +200,53 @@ func (s *ConfigStore) SaveJobTypeConfig(config *plugin_pb.PersistedJobTypeConfig
 	return nil
 }
 
+// SaveJobTypeConfigIfNotExists atomically checks whether a config for the
+// given job type already exists and only persists config when none is found.
+// Returns true if the config was saved, false if a config already existed.
+func (s *ConfigStore) SaveJobTypeConfigIfNotExists(config *plugin_pb.PersistedJobTypeConfig) (bool, error) {
+	if config == nil {
+		return false, fmt.Errorf("job type config is nil")
+	}
+	if config.JobType == "" {
+		return false, fmt.Errorf("job type config has empty job_type")
+	}
+	sanitizedJobType, err := sanitizeJobType(config.JobType)
+	if err != nil {
+		return false, err
+	}
+	config.JobType = sanitizedJobType
+
+	clone := proto.Clone(config).(*plugin_pb.PersistedJobTypeConfig)
+
+	s.mu.Lock()
+	defer s.mu.Unlock()
+
+	if !s.configured {
+		if _, exists := s.memConfigs[config.JobType]; exists {
+			return false, nil
+		}
+		s.memConfigs[config.JobType] = clone
+		return true, nil
+	}
+
+	pbPath := filepath.Join(s.baseDir, jobTypesDirName, config.JobType, configPBFileName)
+	if _, statErr := os.Stat(pbPath); statErr == nil {
+		return false, nil
+	}
+
+	jobTypeDir, err := s.ensureJobTypeDir(config.JobType)
+	if err != nil {
+		return false, err
+	}
+
+	jsonPath := filepath.Join(jobTypeDir, configJSONFileName)
+	if err := writeProtoFiles(clone, filepath.Join(jobTypeDir, configPBFileName), jsonPath); err != nil {
+		return false, fmt.Errorf("save job type config for %s: %w", config.JobType, err)
+	}
+
+	return true, nil
+}
+
 func (s *ConfigStore) LoadJobTypeConfig(jobType string) (*plugin_pb.PersistedJobTypeConfig, error) {
 	if _, err := sanitizeJobType(jobType); err != nil {
 		return nil, err

weed/admin/plugin/config_store_test.go

@@ -208,6 +208,81 @@ func TestConfigStoreMonitorStateRoundTrip(t *testing.T) {
 	}
 }
 
+func TestConfigStoreSaveJobTypeConfigIfNotExists(t *testing.T) {
+	t.Parallel()
+
+	t.Run("in-memory", func(t *testing.T) {
+		t.Parallel()
+		store, err := NewConfigStore("")
+		if err != nil {
+			t.Fatalf("NewConfigStore: %v", err)
+		}
+		testSaveJobTypeConfigIfNotExists(t, store)
+	})
+
+	t.Run("on-disk", func(t *testing.T) {
+		t.Parallel()
+		store, err := NewConfigStore(t.TempDir())
+		if err != nil {
+			t.Fatalf("NewConfigStore: %v", err)
+		}
+		testSaveJobTypeConfigIfNotExists(t, store)
+	})
+}
+
+func testSaveJobTypeConfigIfNotExists(t *testing.T, store *ConfigStore) {
+	t.Helper()
+
+	cfg := &plugin_pb.PersistedJobTypeConfig{
+		JobType:      "admin_script",
+		AdminRuntime: &plugin_pb.AdminRuntimeConfig{Enabled: true},
+	}
+
+	// First call should save.
+	saved, err := store.SaveJobTypeConfigIfNotExists(cfg)
+	if err != nil {
+		t.Fatalf("first SaveJobTypeConfigIfNotExists: %v", err)
+	}
+	if !saved {
+		t.Fatal("expected first call to save the config")
+	}
+
+	// Second call with same job type should not save.
+	saved, err = store.SaveJobTypeConfigIfNotExists(&plugin_pb.PersistedJobTypeConfig{
+		JobType:      "admin_script",
+		AdminRuntime: &plugin_pb.AdminRuntimeConfig{Enabled: false},
+	})
+	if err != nil {
+		t.Fatalf("second SaveJobTypeConfigIfNotExists: %v", err)
+	}
+	if saved {
+		t.Fatal("expected second call to be a no-op")
+	}
+
+	// Verify the original config was preserved.
+	loaded, err := store.LoadJobTypeConfig("admin_script")
+	if err != nil {
+		t.Fatalf("LoadJobTypeConfig: %v", err)
+	}
+	if loaded == nil {
+		t.Fatal("expected config to exist")
+	}
+	if !loaded.AdminRuntime.Enabled {
+		t.Fatal("expected original config (Enabled=true) to be preserved")
+	}
+
+	// Different job type should still save.
+	saved, err = store.SaveJobTypeConfigIfNotExists(&plugin_pb.PersistedJobTypeConfig{
+		JobType: "vacuum",
+	})
+	if err != nil {
+		t.Fatalf("SaveJobTypeConfigIfNotExists for different type: %v", err)
+	}
+	if !saved {
+		t.Fatal("expected save for a different job type")
+	}
+}
+
 func TestConfigStoreJobDetailRoundTrip(t *testing.T) {
 	t.Parallel()
 

weed/admin/plugin/plugin.go

@@ -34,6 +34,11 @@ type Options struct {
 	SchedulerTick          time.Duration
 	ClusterContextProvider func(context.Context) (*plugin_pb.ClusterContext, error)
 	LockManager            LockManager
+	// ConfigDefaultsProvider is an optional callback invoked when a job type's
+	// config is being bootstrapped from its descriptor defaults. It can enrich
+	// or replace the default config before it is persisted. If nil, descriptor
+	// defaults are used as-is.
+	ConfigDefaultsProvider func(config *plugin_pb.PersistedJobTypeConfig) *plugin_pb.PersistedJobTypeConfig
 }
 
 // JobTypeInfo contains metadata about a plugin job type.
@@ -54,6 +59,7 @@ type Plugin struct {
 
 	schedulerTick          time.Duration
 	clusterContextProvider func(context.Context) (*plugin_pb.ClusterContext, error)
+	configDefaultsProvider func(config *plugin_pb.PersistedJobTypeConfig) *plugin_pb.PersistedJobTypeConfig
 	lockManager            LockManager
 
 	schedulerMu       sync.Mutex
@@ -68,6 +74,11 @@ type Plugin struct {
 	adminScriptRunMu          sync.RWMutex
 	schedulerDetectionMu      sync.Mutex
 	schedulerDetection        map[string]*schedulerDetectionInfo
+	schedulerRunMu            sync.Mutex
+	schedulerRun              map[string]*schedulerRunInfo
+	schedulerLoopMu           sync.Mutex
+	schedulerLoopState        schedulerLoopState
+	schedulerWakeCh chan struct{}
 
 	dedupeMu           sync.Mutex
 	recentDedupeByType map[string]map[string]time.Time
@@ -154,6 +165,7 @@ func New(options Options) (*Plugin, error) {
 		sendTimeout:               sendTimeout,
 		schedulerTick:             schedulerTick,
 		clusterContextProvider:    options.ClusterContextProvider,
+		configDefaultsProvider:    options.ConfigDefaultsProvider,
 		lockManager:               options.LockManager,
 		sessions:                  make(map[string]*streamSession),
 		pendingSchema:             make(map[string]chan *plugin_pb.ConfigSchemaResponse),
@@ -164,10 +176,12 @@ func New(options Options) (*Plugin, error) {
 		detectorLeases:            make(map[string]string),
 		schedulerExecReservations: make(map[string]int),
 		schedulerDetection:        make(map[string]*schedulerDetectionInfo),
+		schedulerRun:              make(map[string]*schedulerRunInfo),
 		recentDedupeByType:        make(map[string]map[string]time.Time),
 		jobs:                      make(map[string]*TrackedJob),
 		activities:                make([]JobActivity, 0, 256),
 		persistTicker:             time.NewTicker(2 * time.Second),
+		schedulerWakeCh:           make(chan struct{}, 1),
 		shutdownCh:                make(chan struct{}),
 	}
 	plugin.ctx, plugin.ctxCancel = context.WithCancel(context.Background())
@@ -371,9 +385,14 @@ func (r *Plugin) LoadJobTypeConfig(jobType string) (*plugin_pb.PersistedJobTypeC
 }
 
 func (r *Plugin) SaveJobTypeConfig(config *plugin_pb.PersistedJobTypeConfig) error {
-	return r.store.SaveJobTypeConfig(config)
+	if err := r.store.SaveJobTypeConfig(config); err != nil {
+		return err
+	}
+	r.wakeScheduler()
+	return nil
 }
 
+
 func (r *Plugin) LoadDescriptor(jobType string) (*plugin_pb.JobTypeDescriptor, error) {
 	return r.store.LoadDescriptor(jobType)
 }
@@ -912,6 +931,7 @@ func (r *Plugin) handleWorkerMessage(workerID string, message *plugin_pb.WorkerT
 	switch body := message.Body.(type) {
 	case *plugin_pb.WorkerToAdminMessage_Hello:
 		r.registry.UpsertFromHello(body.Hello)
+		r.wakeScheduler()
 	case *plugin_pb.WorkerToAdminMessage_Heartbeat:
 		r.registry.UpdateHeartbeat(workerID, body.Heartbeat)
 	case *plugin_pb.WorkerToAdminMessage_ConfigSchemaResponse:
@@ -981,14 +1001,6 @@ func (r *Plugin) ensureJobTypeConfigFromDescriptor(jobType string, descriptor *p
 		return nil
 	}
 
-	existing, err := r.store.LoadJobTypeConfig(jobType)
-	if err != nil {
-		return err
-	}
-	if existing != nil {
-		return nil
-	}
-
 	workerDefaults := CloneConfigValueMap(descriptor.WorkerDefaultValues)
 	if len(workerDefaults) == 0 && descriptor.WorkerConfigForm != nil {
 		workerDefaults = CloneConfigValueMap(descriptor.WorkerConfigForm.DefaultValues)
@@ -1011,6 +1023,7 @@ func (r *Plugin) ensureJobTypeConfigFromDescriptor(jobType string, descriptor *p
 			PerWorkerExecutionConcurrency: defaults.PerWorkerExecutionConcurrency,
 			RetryLimit:                    defaults.RetryLimit,
 			RetryBackoffSeconds:           defaults.RetryBackoffSeconds,
+			JobTypeMaxRuntimeSeconds:      defaults.JobTypeMaxRuntimeSeconds,
 		}
 	}
 
@@ -1024,7 +1037,22 @@ func (r *Plugin) ensureJobTypeConfigFromDescriptor(jobType string, descriptor *p
 		UpdatedBy:          "plugin",
 	}
 
-	return r.store.SaveJobTypeConfig(cfg)
+	// Check existence first to avoid calling configDefaultsProvider unnecessarily
+	// (e.g., it may make a blocking gRPC call to fetch master config).
+	existing, err := r.store.LoadJobTypeConfig(jobType)
+	if err != nil {
+		return err
+	}
+	if existing != nil {
+		return nil
+	}
+
+	if r.configDefaultsProvider != nil {
+		cfg = r.configDefaultsProvider(cfg)
+	}
+
+	_, err = r.store.SaveJobTypeConfigIfNotExists(cfg)
+	return err
 }
 
 func (r *Plugin) handleDetectionProposals(workerID string, message *plugin_pb.DetectionProposals) {

weed/admin/plugin/plugin_monitor.go

@@ -861,6 +861,79 @@ func (r *Plugin) trackExecutionQueued(job *plugin_pb.JobSpec) {
 	})
 }
 
+func (r *Plugin) cancelQueuedJob(job *plugin_pb.JobSpec, cause error) {
+	reason := "job canceled"
+	if cause != nil {
+		reason = cause.Error()
+	}
+	r.markJobCanceled(job, reason)
+}
+
+func (r *Plugin) markJobCanceled(job *plugin_pb.JobSpec, reason string) {
+	if job == nil || strings.TrimSpace(job.JobId) == "" {
+		return
+	}
+
+	now := time.Now().UTC()
+	if strings.TrimSpace(reason) == "" {
+		reason = "job canceled"
+	}
+
+	r.jobsMu.Lock()
+	tracked := r.jobs[job.JobId]
+	if tracked == nil {
+		tracked = &TrackedJob{
+			JobID:     job.JobId,
+			CreatedAt: timeToPtr(now),
+		}
+		r.jobs[job.JobId] = tracked
+	}
+
+	if job.JobType != "" {
+		tracked.JobType = job.JobType
+	}
+	tracked.State = StateCanceled
+	tracked.Stage = "canceled"
+	tracked.Message = reason
+	tracked.ErrorMessage = reason
+	tracked.Progress = 0
+	if tracked.CreatedAt == nil || tracked.CreatedAt.IsZero() {
+		tracked.CreatedAt = timeToPtr(now)
+	}
+	tracked.UpdatedAt = timeToPtr(now)
+	tracked.CompletedAt = timeToPtr(now)
+	trackedSnapshot := cloneTrackedJob(*tracked)
+	r.pruneTrackedJobsLocked()
+	r.dirtyJobs = true
+	r.jobsMu.Unlock()
+
+	r.persistJobDetailSnapshot(job.JobId, func(detail *TrackedJob) {
+		detail.JobID = job.JobId
+		if job.JobType != "" {
+			detail.JobType = job.JobType
+		}
+		detail.State = trackedSnapshot.State
+		detail.Stage = trackedSnapshot.Stage
+		detail.Message = trackedSnapshot.Message
+		detail.ErrorMessage = trackedSnapshot.ErrorMessage
+		detail.Progress = trackedSnapshot.Progress
+		if detail.CreatedAt == nil || detail.CreatedAt.IsZero() {
+			detail.CreatedAt = trackedSnapshot.CreatedAt
+		}
+		detail.UpdatedAt = trackedSnapshot.UpdatedAt
+		detail.CompletedAt = trackedSnapshot.CompletedAt
+	})
+
+	r.appendActivity(JobActivity{
+		JobID:      job.JobId,
+		JobType:    job.JobType,
+		Source:     "admin_scheduler",
+		Message:    reason,
+		Stage:      "canceled",
+		OccurredAt: timeToPtr(now),
+	})
+}
+
 func (r *Plugin) trackExecutionCompletion(completed *plugin_pb.JobCompleted) *TrackedJob {
 	if completed == nil || strings.TrimSpace(completed.JobId) == "" {
 		return nil

weed/admin/plugin/plugin_scheduler.go

@@ -13,13 +13,17 @@ import (
 	"google.golang.org/protobuf/types/known/timestamppb"
 )
 
-var errExecutorAtCapacity = errors.New("executor is at capacity")
+var (
+	errExecutorAtCapacity = errors.New("executor is at capacity")
+	errSchedulerShutdown  = errors.New("scheduler shutdown")
+)
 
 const (
 	defaultSchedulerTick                       = 5 * time.Second
 	defaultScheduledDetectionInterval          = 300 * time.Second
 	defaultScheduledDetectionTimeout           = 45 * time.Second
 	defaultScheduledExecutionTimeout           = 90 * time.Second
+	defaultScheduledJobTypeMaxRuntime          = 30 * time.Minute
 	defaultScheduledMaxResults           int32 = 1000
 	defaultScheduledExecutionConcurrency       = 1
 	defaultScheduledPerWorkerConcurrency       = 1
@@ -28,12 +32,14 @@ const (
 	defaultClusterContextTimeout               = 10 * time.Second
 	defaultWaitingBacklogFloor                 = 8
 	defaultWaitingBacklogMultiplier            = 4
+	maxEstimatedRuntimeCap                     = 8 * time.Hour
 )
 
 type schedulerPolicy struct {
 	DetectionInterval      time.Duration
 	DetectionTimeout       time.Duration
 	ExecutionTimeout       time.Duration
+	JobTypeMaxRuntime      time.Duration
 	RetryBackoff           time.Duration
 	MaxResults             int32
 	ExecutionConcurrency   int
@@ -44,31 +50,71 @@ type schedulerPolicy struct {
 
 func (r *Plugin) schedulerLoop() {
 	defer r.wg.Done()
-	ticker := time.NewTicker(r.schedulerTick)
-	defer ticker.Stop()
-
-	// Try once immediately on startup.
-	r.runSchedulerTick()
-
 	for {
 		select {
 		case <-r.shutdownCh:
 			return
-		case <-ticker.C:
-			r.runSchedulerTick()
+		default:
+		}
+
+		hadJobs := r.runSchedulerIteration()
+		r.recordSchedulerIterationComplete(hadJobs)
+
+		if hadJobs {
+			continue
+		}
+
+		r.setSchedulerLoopState("", "sleeping")
+		idleSleep := defaultSchedulerIdleSleep
+		if nextRun := r.earliestNextDetectionAt(); !nextRun.IsZero() {
+			if until := time.Until(nextRun); until <= 0 {
+				idleSleep = 0
+			} else if until < idleSleep {
+				idleSleep = until
+			}
+		}
+		if idleSleep <= 0 {
+			continue
+		}
+
+		timer := time.NewTimer(idleSleep)
+		select {
+		case <-r.shutdownCh:
+			timer.Stop()
+			return
+		case <-r.schedulerWakeCh:
+			if !timer.Stop() {
+				<-timer.C
+			}
+			continue
+		case <-timer.C:
 		}
 	}
 }
 
-func (r *Plugin) runSchedulerTick() {
+func (r *Plugin) runSchedulerIteration() bool {
 	r.expireStaleJobs(time.Now().UTC())
 
 	jobTypes := r.registry.DetectableJobTypes()
 	if len(jobTypes) == 0 {
-		return
+		r.setSchedulerLoopState("", "idle")
+		return false
+	}
+
+	r.setSchedulerLoopState("", "waiting_for_lock")
+	releaseLock, err := r.acquireAdminLock("plugin scheduler iteration")
+	if err != nil {
+		glog.Warningf("Plugin scheduler failed to acquire lock: %v", err)
+		r.setSchedulerLoopState("", "idle")
+		return false
+	}
+	if releaseLock != nil {
+		defer releaseLock()
 	}
 
 	active := make(map[string]struct{}, len(jobTypes))
+	hadJobs := false
+
 	for _, jobType := range jobTypes {
 		active[jobType] = struct{}{}
 
@@ -81,20 +127,243 @@ func (r *Plugin) runSchedulerTick() {
 			r.clearSchedulerJobType(jobType)
 			continue
 		}
-
-		if !r.markDetectionDue(jobType, policy.DetectionInterval) {
+		initialDelay := time.Duration(0)
+		if runInfo := r.snapshotSchedulerRun(jobType); runInfo.lastRunStartedAt.IsZero() {
+			initialDelay = 5 * time.Second
+		}
+		if !r.markDetectionDue(jobType, policy.DetectionInterval, initialDelay) {
 			continue
 		}
 
-		r.wg.Add(1)
-		go func(jt string, p schedulerPolicy) {
-			defer r.wg.Done()
-			r.runScheduledDetection(jt, p)
-		}(jobType, policy)
+		detected := r.runJobTypeIteration(jobType, policy)
+		if detected {
+			hadJobs = true
+		}
 	}
 
 	r.pruneSchedulerState(active)
 	r.pruneDetectorLeases(active)
+	r.setSchedulerLoopState("", "idle")
+	return hadJobs
+}
+
+func (r *Plugin) wakeScheduler() {
+	if r == nil {
+		return
+	}
+	select {
+	case r.schedulerWakeCh <- struct{}{}:
+	default:
+	}
+}
+
+func (r *Plugin) runJobTypeIteration(jobType string, policy schedulerPolicy) bool {
+	r.recordSchedulerRunStart(jobType)
+	r.clearWaitingJobQueue(jobType)
+	r.setSchedulerLoopState(jobType, "detecting")
+	r.markJobTypeInFlight(jobType)
+	defer r.finishDetection(jobType)
+
+	start := time.Now().UTC()
+	r.appendActivity(JobActivity{
+		JobType:    jobType,
+		Source:     "admin_scheduler",
+		Message:    "scheduled detection started",
+		Stage:      "detecting",
+		OccurredAt: timeToPtr(start),
+	})
+
+	if skip, waitingCount, waitingThreshold := r.shouldSkipDetectionForWaitingJobs(jobType, policy); skip {
+		r.recordSchedulerDetectionSkip(jobType, fmt.Sprintf("waiting backlog %d reached threshold %d", waitingCount, waitingThreshold))
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    fmt.Sprintf("scheduled detection skipped: waiting backlog %d reached threshold %d", waitingCount, waitingThreshold),
+			Stage:      "skipped_waiting_backlog",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+		r.recordSchedulerRunComplete(jobType, "skipped")
+		return false
+	}
+
+	maxRuntime := policy.JobTypeMaxRuntime
+	if maxRuntime <= 0 {
+		maxRuntime = defaultScheduledJobTypeMaxRuntime
+	}
+	jobCtx, cancel := context.WithTimeout(context.Background(), maxRuntime)
+	defer cancel()
+
+	clusterContext, err := r.loadSchedulerClusterContext(jobCtx)
+	if err != nil {
+		r.recordSchedulerDetectionError(jobType, err)
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    fmt.Sprintf("scheduled detection aborted: %v", err),
+			Stage:      "failed",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+		r.recordSchedulerRunComplete(jobType, "error")
+		return false
+	}
+
+	detectionTimeout := policy.DetectionTimeout
+	remaining := time.Until(start.Add(maxRuntime))
+	if remaining <= 0 {
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    "scheduled run timed out before detection",
+			Stage:      "timeout",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+		r.recordSchedulerRunComplete(jobType, "timeout")
+		return false
+	}
+	if detectionTimeout <= 0 {
+		detectionTimeout = defaultScheduledDetectionTimeout
+	}
+	if detectionTimeout > remaining {
+		detectionTimeout = remaining
+	}
+
+	detectCtx, cancelDetect := context.WithTimeout(jobCtx, detectionTimeout)
+	proposals, err := r.RunDetection(detectCtx, jobType, clusterContext, policy.MaxResults)
+	cancelDetect()
+	if err != nil {
+		r.recordSchedulerDetectionError(jobType, err)
+		stage := "failed"
+		status := "error"
+		if errors.Is(err, context.DeadlineExceeded) || errors.Is(err, context.Canceled) {
+			stage = "timeout"
+			status = "timeout"
+		}
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    fmt.Sprintf("scheduled detection failed: %v", err),
+			Stage:      stage,
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+		r.recordSchedulerRunComplete(jobType, status)
+		return false
+	}
+
+	r.appendActivity(JobActivity{
+		JobType:    jobType,
+		Source:     "admin_scheduler",
+		Message:    fmt.Sprintf("scheduled detection completed: %d proposal(s)", len(proposals)),
+		Stage:      "detected",
+		OccurredAt: timeToPtr(time.Now().UTC()),
+	})
+	r.recordSchedulerDetectionSuccess(jobType, len(proposals))
+
+	detected := len(proposals) > 0
+
+	filteredByActive, skippedActive := r.filterProposalsWithActiveJobs(jobType, proposals)
+	if skippedActive > 0 {
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    fmt.Sprintf("scheduled detection skipped %d proposal(s) due to active assigned/running jobs", skippedActive),
+			Stage:      "deduped_active_jobs",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+	}
+
+	if len(filteredByActive) == 0 {
+		r.recordSchedulerRunComplete(jobType, "success")
+		return detected
+	}
+
+	filtered := r.filterScheduledProposals(filteredByActive)
+	if len(filtered) != len(filteredByActive) {
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    fmt.Sprintf("scheduled detection deduped %d proposal(s) within this run", len(filteredByActive)-len(filtered)),
+			Stage:      "deduped",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+	}
+
+	if len(filtered) == 0 {
+		r.recordSchedulerRunComplete(jobType, "success")
+		return detected
+	}
+
+	r.setSchedulerLoopState(jobType, "executing")
+
+	// Scan proposals for the maximum estimated_runtime_seconds so the
+	// execution phase gets enough time for large jobs (e.g. vacuum on
+	// big volumes). If any proposal needs more time than the remaining
+	// JobTypeMaxRuntime, extend the execution context accordingly.
+	var maxEstimatedRuntime time.Duration
+	for _, p := range filtered {
+		if p.Parameters != nil {
+			if est, ok := p.Parameters["estimated_runtime_seconds"]; ok {
+				if v := est.GetInt64Value(); v > 0 {
+					if d := time.Duration(v) * time.Second; d > maxEstimatedRuntime {
+						maxEstimatedRuntime = d
+					}
+				}
+			}
+		}
+	}
+	if maxEstimatedRuntime > maxEstimatedRuntimeCap {
+		maxEstimatedRuntime = maxEstimatedRuntimeCap
+	}
+
+	remaining = time.Until(start.Add(maxRuntime))
+	if remaining <= 0 {
+		r.appendActivity(JobActivity{
+			JobType:    jobType,
+			Source:     "admin_scheduler",
+			Message:    "scheduled execution skipped: job type max runtime reached",
+			Stage:      "timeout",
+			OccurredAt: timeToPtr(time.Now().UTC()),
+		})
+		r.recordSchedulerRunComplete(jobType, "timeout")
+		return detected
+	}
+
+	// If the longest estimated job exceeds the remaining JobTypeMaxRuntime,
+	// create a new execution context with enough headroom instead of using
+	// jobCtx which would cancel too early.
+	execCtx := jobCtx
+	execCancel := context.CancelFunc(func() {})
+	if maxEstimatedRuntime > 0 && maxEstimatedRuntime > remaining {
+		execCtx, execCancel = context.WithTimeout(context.Background(), maxEstimatedRuntime)
+		remaining = maxEstimatedRuntime
+	}
+	defer execCancel()
+
+	execPolicy := policy
+	if execPolicy.ExecutionTimeout <= 0 {
+		execPolicy.ExecutionTimeout = defaultScheduledExecutionTimeout
+	}
+	if execPolicy.ExecutionTimeout > remaining {
+		execPolicy.ExecutionTimeout = remaining
+	}
+
+	successCount, errorCount, canceledCount := r.dispatchScheduledProposals(execCtx, jobType, filtered, clusterContext, execPolicy)
+
+	status := "success"
+	if execCtx.Err() != nil {
+		status = "timeout"
+	} else if errorCount > 0 || canceledCount > 0 {
+		status = "error"
+	}
+
+	r.appendActivity(JobActivity{
+		JobType:    jobType,
+		Source:     "admin_scheduler",
+		Message:    fmt.Sprintf("scheduled execution finished: success=%d error=%d canceled=%d", successCount, errorCount, canceledCount),
+		Stage:      "executed",
+		OccurredAt: timeToPtr(time.Now().UTC()),
+	})
+	r.recordSchedulerRunComplete(jobType, status)
+	return detected
 }
 
 func (r *Plugin) loadSchedulerPolicy(jobType string) (schedulerPolicy, bool, error) {
@@ -119,6 +388,7 @@ func (r *Plugin) loadSchedulerPolicy(jobType string) (schedulerPolicy, bool, err
 		DetectionInterval:      durationFromSeconds(adminRuntime.DetectionIntervalSeconds, defaultScheduledDetectionInterval),
 		DetectionTimeout:       durationFromSeconds(adminRuntime.DetectionTimeoutSeconds, defaultScheduledDetectionTimeout),
 		ExecutionTimeout:       defaultScheduledExecutionTimeout,
+		JobTypeMaxRuntime:      durationFromSeconds(adminRuntime.JobTypeMaxRuntimeSeconds, defaultScheduledJobTypeMaxRuntime),
 		RetryBackoff:           durationFromSeconds(adminRuntime.RetryBackoffSeconds, defaultScheduledRetryBackoff),
 		MaxResults:             adminRuntime.MaxJobsPerDetection,
 		ExecutionConcurrency:   int(adminRuntime.GlobalExecutionConcurrency),
@@ -148,6 +418,9 @@ func (r *Plugin) loadSchedulerPolicy(jobType string) (schedulerPolicy, bool, err
 	if policy.RetryLimit < 0 {
 		policy.RetryLimit = 0
 	}
+	if policy.JobTypeMaxRuntime <= 0 {
+		policy.JobTypeMaxRuntime = defaultScheduledJobTypeMaxRuntime
+	}
 
 	// Plugin protocol currently has only detection timeout in admin settings.
 	execTimeout := time.Duration(adminRuntime.DetectionTimeoutSeconds*2) * time.Second
@@ -199,6 +472,7 @@ func (r *Plugin) ListSchedulerStates() ([]SchedulerJobTypeState, error) {
 				state.DetectionIntervalSeconds = secondsFromDuration(policy.DetectionInterval)
 				state.DetectionTimeoutSeconds = secondsFromDuration(policy.DetectionTimeout)
 				state.ExecutionTimeoutSeconds = secondsFromDuration(policy.ExecutionTimeout)
+				state.JobTypeMaxRuntimeSeconds = secondsFromDuration(policy.JobTypeMaxRuntime)
 				state.MaxJobsPerDetection = policy.MaxResults
 				state.GlobalExecutionConcurrency = policy.ExecutionConcurrency
 				state.PerWorkerExecutionConcurrency = policy.PerWorkerConcurrency
@@ -207,6 +481,19 @@ func (r *Plugin) ListSchedulerStates() ([]SchedulerJobTypeState, error) {
 			}
 		}
 
+		runInfo := r.snapshotSchedulerRun(jobType)
+		if !runInfo.lastRunStartedAt.IsZero() {
+			at := runInfo.lastRunStartedAt
+			state.LastRunStartedAt = &at
+		}
+		if !runInfo.lastRunCompletedAt.IsZero() {
+			at := runInfo.lastRunCompletedAt
+			state.LastRunCompletedAt = &at
+		}
+		if runInfo.lastRunStatus != "" {
+			state.LastRunStatus = runInfo.lastRunStatus
+		}
+
 		leasedWorkerID := r.getDetectorLease(jobType)
 		if leasedWorkerID != "" {
 			state.DetectorWorkerID = leasedWorkerID
@@ -258,10 +545,11 @@ func deriveSchedulerAdminRuntime(
 		PerWorkerExecutionConcurrency: defaults.PerWorkerExecutionConcurrency,
 		RetryLimit:                    defaults.RetryLimit,
 		RetryBackoffSeconds:           defaults.RetryBackoffSeconds,
+		JobTypeMaxRuntimeSeconds:      defaults.JobTypeMaxRuntimeSeconds,
 	}
 }
 
-func (r *Plugin) markDetectionDue(jobType string, interval time.Duration) bool {
+func (r *Plugin) markDetectionDue(jobType string, interval, initialDelay time.Duration) bool {
 	now := time.Now().UTC()
 
 	r.schedulerMu.Lock()
@@ -275,12 +563,43 @@ func (r *Plugin) markDetectionDue(jobType string, interval time.Duration) bool {
 	if exists && now.Before(nextRun) {
 		return false
 	}
+	if !exists && initialDelay > 0 {
+		r.nextDetectionAt[jobType] = now.Add(initialDelay)
+		return false
+	}
 
 	r.nextDetectionAt[jobType] = now.Add(interval)
 	r.detectionInFlight[jobType] = true
 	return true
 }
 
+func (r *Plugin) earliestNextDetectionAt() time.Time {
+	if r == nil {
+		return time.Time{}
+	}
+
+	r.schedulerMu.Lock()
+	defer r.schedulerMu.Unlock()
+
+	var earliest time.Time
+	for _, nextRun := range r.nextDetectionAt {
+		if nextRun.IsZero() {
+			continue
+		}
+		if earliest.IsZero() || nextRun.Before(earliest) {
+			earliest = nextRun
+		}
+	}
+
+	return earliest
+}
+
+func (r *Plugin) markJobTypeInFlight(jobType string) {
+	r.schedulerMu.Lock()
+	r.detectionInFlight[jobType] = true
+	r.schedulerMu.Unlock()
+}
+
 func (r *Plugin) finishDetection(jobType string) {
 	r.schedulerMu.Lock()
 	delete(r.detectionInFlight, jobType)
@@ -318,125 +637,18 @@ func (r *Plugin) pruneDetectorLeases(activeJobTypes map[string]struct{}) {
 	}
 }
 
-func (r *Plugin) runScheduledDetection(jobType string, policy schedulerPolicy) {
-	defer r.finishDetection(jobType)
-
-	releaseLock, lockErr := r.acquireAdminLock(fmt.Sprintf("plugin scheduled detection %s", jobType))
-	if lockErr != nil {
-		r.recordSchedulerDetectionError(jobType, lockErr)
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection aborted: failed to acquire lock: %v", lockErr),
-			Stage:      "failed",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-		return
-	}
-	if releaseLock != nil {
-		defer releaseLock()
-	}
-
-	start := time.Now().UTC()
-	r.appendActivity(JobActivity{
-		JobType:    jobType,
-		Source:     "admin_scheduler",
-		Message:    "scheduled detection started",
-		Stage:      "detecting",
-		OccurredAt: timeToPtr(start),
-	})
-
-	if skip, waitingCount, waitingThreshold := r.shouldSkipDetectionForWaitingJobs(jobType, policy); skip {
-		r.recordSchedulerDetectionSkip(jobType, fmt.Sprintf("waiting backlog %d reached threshold %d", waitingCount, waitingThreshold))
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection skipped: waiting backlog %d reached threshold %d", waitingCount, waitingThreshold),
-			Stage:      "skipped_waiting_backlog",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-		return
-	}
-
-	clusterContext, err := r.loadSchedulerClusterContext()
-	if err != nil {
-		r.recordSchedulerDetectionError(jobType, err)
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection aborted: %v", err),
-			Stage:      "failed",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-		return
-	}
-
-	ctx, cancel := context.WithTimeout(context.Background(), policy.DetectionTimeout)
-	proposals, err := r.RunDetection(ctx, jobType, clusterContext, policy.MaxResults)
-	cancel()
-	if err != nil {
-		r.recordSchedulerDetectionError(jobType, err)
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection failed: %v", err),
-			Stage:      "failed",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-		return
-	}
-
-	r.appendActivity(JobActivity{
-		JobType:    jobType,
-		Source:     "admin_scheduler",
-		Message:    fmt.Sprintf("scheduled detection completed: %d proposal(s)", len(proposals)),
-		Stage:      "detected",
-		OccurredAt: timeToPtr(time.Now().UTC()),
-	})
-	r.recordSchedulerDetectionSuccess(jobType, len(proposals))
-
-	filteredByActive, skippedActive := r.filterProposalsWithActiveJobs(jobType, proposals)
-	if skippedActive > 0 {
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection skipped %d proposal(s) due to active assigned/running jobs", skippedActive),
-			Stage:      "deduped_active_jobs",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-	}
-
-	if len(filteredByActive) == 0 {
-		return
-	}
-
-	filtered := r.filterScheduledProposals(filteredByActive)
-	if len(filtered) != len(filteredByActive) {
-		r.appendActivity(JobActivity{
-			JobType:    jobType,
-			Source:     "admin_scheduler",
-			Message:    fmt.Sprintf("scheduled detection deduped %d proposal(s) within this run", len(filteredByActive)-len(filtered)),
-			Stage:      "deduped",
-			OccurredAt: timeToPtr(time.Now().UTC()),
-		})
-	}
-
-	if len(filtered) == 0 {
-		return
-	}
-
-	r.dispatchScheduledProposals(jobType, filtered, clusterContext, policy)
-}
-
-func (r *Plugin) loadSchedulerClusterContext() (*plugin_pb.ClusterContext, error) {
+func (r *Plugin) loadSchedulerClusterContext(ctx context.Context) (*plugin_pb.ClusterContext, error) {
 	if r.clusterContextProvider == nil {
 		return nil, fmt.Errorf("cluster context provider is not configured")
 	}
 
-	ctx, cancel := context.WithTimeout(context.Background(), defaultClusterContextTimeout)
+	if ctx == nil {
+		ctx = context.Background()
+	}
+	clusterCtx, cancel := context.WithTimeout(ctx, defaultClusterContextTimeout)
 	defer cancel()
 
-	clusterContext, err := r.clusterContextProvider(ctx)
+	clusterContext, err := r.clusterContextProvider(clusterCtx)
 	if err != nil {
 		return nil, err
 	}
@@ -447,11 +659,16 @@ func (r *Plugin) loadSchedulerClusterContext() (*plugin_pb.ClusterContext, error
 }
 
 func (r *Plugin) dispatchScheduledProposals(
+	ctx context.Context,
 	jobType string,
 	proposals []*plugin_pb.JobProposal,
 	clusterContext *plugin_pb.ClusterContext,
 	policy schedulerPolicy,
-) {
+) (int, int, int) {
+	if ctx == nil {
+		ctx = context.Background()
+	}
+
 	jobQueue := make(chan *plugin_pb.JobSpec, len(proposals))
 	for index, proposal := range proposals {
 		job := buildScheduledJobSpec(jobType, proposal, index)
@@ -459,7 +676,7 @@ func (r *Plugin) dispatchScheduledProposals(
 		select {
 		case <-r.shutdownCh:
 			close(jobQueue)
-			return
+			return 0, 0, 0
 		default:
 			jobQueue <- job
 		}
@@ -470,6 +687,7 @@ func (r *Plugin) dispatchScheduledProposals(
 	var statsMu sync.Mutex
 	successCount := 0
 	errorCount := 0
+	canceledCount := 0
 
 	workerCount := policy.ExecutionConcurrency
 	if workerCount < 1 {
@@ -481,6 +699,7 @@ func (r *Plugin) dispatchScheduledProposals(
 		go func() {
 			defer wg.Done()
 
+		jobLoop:
 			for job := range jobQueue {
 				select {
 				case <-r.shutdownCh:
@@ -488,19 +707,36 @@ func (r *Plugin) dispatchScheduledProposals(
 				default:
 				}
 
+				if ctx.Err() != nil {
+					r.cancelQueuedJob(job, ctx.Err())
+					statsMu.Lock()
+					canceledCount++
+					statsMu.Unlock()
+					continue
+				}
+
 				for {
 					select {
 					case <-r.shutdownCh:
 						return
 					default:
 					}
+					if ctx.Err() != nil {
+						r.cancelQueuedJob(job, ctx.Err())
+						statsMu.Lock()
+						canceledCount++
+						statsMu.Unlock()
+						continue jobLoop
+					}
 
-					executor, release, reserveErr := r.reserveScheduledExecutor(jobType, policy)
+					executor, release, reserveErr := r.reserveScheduledExecutor(ctx, jobType, policy)
 					if reserveErr != nil {
-						select {
-						case <-r.shutdownCh:
-							return
-						default:
+						if ctx.Err() != nil {
+							r.cancelQueuedJob(job, ctx.Err())
+							statsMu.Lock()
+							canceledCount++
+							statsMu.Unlock()
+							continue jobLoop
 						}
 						statsMu.Lock()
 						errorCount++
@@ -515,16 +751,23 @@ func (r *Plugin) dispatchScheduledProposals(
 						break
 					}
 
-					err := r.executeScheduledJobWithExecutor(executor, job, clusterContext, policy)
+					err := r.executeScheduledJobWithExecutor(ctx, executor, job, clusterContext, policy)
 					release()
 					if errors.Is(err, errExecutorAtCapacity) {
 						r.trackExecutionQueued(job)
-						if !waitForShutdownOrTimer(r.shutdownCh, policy.ExecutorReserveBackoff) {
+						if !waitForShutdownOrTimerWithContext(r.shutdownCh, ctx, policy.ExecutorReserveBackoff) {
 							return
 						}
 						continue
 					}
 					if err != nil {
+						if ctx.Err() != nil || errors.Is(err, context.DeadlineExceeded) || errors.Is(err, context.Canceled) {
+							r.cancelQueuedJob(job, err)
+							statsMu.Lock()
+							canceledCount++
+							statsMu.Unlock()
+							continue jobLoop
+						}
 						statsMu.Lock()
 						errorCount++
 						statsMu.Unlock()
@@ -550,23 +793,34 @@ func (r *Plugin) dispatchScheduledProposals(
 
 	wg.Wait()
 
-	r.appendActivity(JobActivity{
-		JobType:    jobType,
-		Source:     "admin_scheduler",
-		Message:    fmt.Sprintf("scheduled execution finished: success=%d error=%d", successCount, errorCount),
-		Stage:      "executed",
-		OccurredAt: timeToPtr(time.Now().UTC()),
-	})
+	drainErr := ctx.Err()
+	if drainErr == nil {
+		drainErr = errSchedulerShutdown
+	}
+	for job := range jobQueue {
+		r.cancelQueuedJob(job, drainErr)
+		canceledCount++
+	}
+
+	return successCount, errorCount, canceledCount
 }
 
 func (r *Plugin) reserveScheduledExecutor(
+	ctx context.Context,
 	jobType string,
 	policy schedulerPolicy,
 ) (*WorkerSession, func(), error) {
+	if ctx == nil {
+		ctx = context.Background()
+	}
+
 	deadline := time.Now().Add(policy.ExecutionTimeout)
 	if policy.ExecutionTimeout <= 0 {
 		deadline = time.Now().Add(10 * time.Minute) // Default cap
 	}
+	if ctxDeadline, ok := ctx.Deadline(); ok && ctxDeadline.Before(deadline) {
+		deadline = ctxDeadline
+	}
 
 	for {
 		select {
@@ -574,6 +828,9 @@ func (r *Plugin) reserveScheduledExecutor(
 			return nil, nil, fmt.Errorf("plugin is shutting down")
 		default:
 		}
+		if ctx.Err() != nil {
+			return nil, nil, ctx.Err()
+		}
 
 		if time.Now().After(deadline) {
 			return nil, nil, fmt.Errorf("timed out waiting for executor capacity for %s", jobType)
@@ -581,7 +838,10 @@ func (r *Plugin) reserveScheduledExecutor(
 
 		executors, err := r.registry.ListExecutors(jobType)
 		if err != nil {
-			if !waitForShutdownOrTimer(r.shutdownCh, policy.ExecutorReserveBackoff) {
+			if !waitForShutdownOrTimerWithContext(r.shutdownCh, ctx, policy.ExecutorReserveBackoff) {
+				if ctx.Err() != nil {
+					return nil, nil, ctx.Err()
+				}
 				return nil, nil, fmt.Errorf("plugin is shutting down")
 			}
 			continue
@@ -595,7 +855,10 @@ func (r *Plugin) reserveScheduledExecutor(
 			return executor, release, nil
 		}
 
-		if !waitForShutdownOrTimer(r.shutdownCh, policy.ExecutorReserveBackoff) {
+		if !waitForShutdownOrTimerWithContext(r.shutdownCh, ctx, policy.ExecutorReserveBackoff) {
+			if ctx.Err() != nil {
+				return nil, nil, ctx.Err()
+			}
 			return nil, nil, fmt.Errorf("plugin is shutting down")
 		}
 	}
@@ -680,6 +943,7 @@ func schedulerWorkerExecutionLimit(executor *WorkerSession, jobType string, poli
 }
 
 func (r *Plugin) executeScheduledJobWithExecutor(
+	ctx context.Context,
 	executor *WorkerSession,
 	job *plugin_pb.JobSpec,
 	clusterContext *plugin_pb.ClusterContext,
@@ -697,8 +961,32 @@ func (r *Plugin) executeScheduledJobWithExecutor(
 			return fmt.Errorf("plugin is shutting down")
 		default:
 		}
+		if ctx != nil && ctx.Err() != nil {
+			return ctx.Err()
+		}
 
-		execCtx, cancel := context.WithTimeout(context.Background(), policy.ExecutionTimeout)
+		parent := ctx
+		if parent == nil {
+			parent = context.Background()
+		}
+		// Use the job's estimated runtime if provided and larger than the
+		// default execution timeout. This lets handlers like vacuum scale
+		// the timeout based on volume size so large volumes are not killed.
+		timeout := policy.ExecutionTimeout
+		if job.Parameters != nil {
+			if est, ok := job.Parameters["estimated_runtime_seconds"]; ok {
+				if v := est.GetInt64Value(); v > 0 {
+					estimated := time.Duration(v) * time.Second
+					if estimated > maxEstimatedRuntimeCap {
+						estimated = maxEstimatedRuntimeCap
+					}
+					if estimated > timeout {
+						timeout = estimated
+					}
+				}
+			}
+		}
+		execCtx, cancel := context.WithTimeout(parent, timeout)
 		_, err := r.executeJobWithExecutor(execCtx, executor, job, clusterContext, int32(attempt))
 		cancel()
 		if err == nil {
@@ -718,7 +1006,10 @@ func (r *Plugin) executeScheduledJobWithExecutor(
 				Stage:      "retry",
 				OccurredAt: timeToPtr(time.Now().UTC()),
 			})
-			if !waitForShutdownOrTimer(r.shutdownCh, policy.RetryBackoff) {
+			if !waitForShutdownOrTimerWithContext(r.shutdownCh, ctx, policy.RetryBackoff) {
+				if ctx != nil && ctx.Err() != nil {
+					return ctx.Err()
+				}
 				return fmt.Errorf("plugin is shutting down")
 			}
 		}
@@ -764,6 +1055,53 @@ func (r *Plugin) countWaitingTrackedJobs(jobType string) int {
 	return waiting
 }
 
+func (r *Plugin) clearWaitingJobQueue(jobType string) int {
+	normalizedJobType := strings.TrimSpace(jobType)
+	if normalizedJobType == "" {
+		return 0
+	}
+
+	jobIDs := make([]string, 0)
+	seen := make(map[string]struct{})
+
+	r.jobsMu.RLock()
+	for _, job := range r.jobs {
+		if job == nil {
+			continue
+		}
+		if strings.TrimSpace(job.JobType) != normalizedJobType {
+			continue
+		}
+		if !isWaitingTrackedJobState(job.State) {
+			continue
+		}
+		jobID := strings.TrimSpace(job.JobID)
+		if jobID == "" {
+			continue
+		}
+		if _, ok := seen[jobID]; ok {
+			continue
+		}
+		seen[jobID] = struct{}{}
+		jobIDs = append(jobIDs, jobID)
+	}
+	r.jobsMu.RUnlock()
+
+	if len(jobIDs) == 0 {
+		return 0
+	}
+
+	reason := fmt.Sprintf("cleared queued job before %s run", normalizedJobType)
+	for _, jobID := range jobIDs {
+		r.markJobCanceled(&plugin_pb.JobSpec{
+			JobId:   jobID,
+			JobType: normalizedJobType,
+		}, reason)
+	}
+
+	return len(jobIDs)
+}
+
 func waitingBacklogThreshold(policy schedulerPolicy) int {
 	concurrency := policy.ExecutionConcurrency
 	if concurrency <= 0 {
@@ -845,10 +1183,13 @@ func secondsFromDuration(duration time.Duration) int32 {
 	return int32(duration / time.Second)
 }
 
-func waitForShutdownOrTimer(shutdown <-chan struct{}, duration time.Duration) bool {
+func waitForShutdownOrTimerWithContext(shutdown <-chan struct{}, ctx context.Context, duration time.Duration) bool {
 	if duration <= 0 {
 		return true
 	}
+	if ctx == nil {
+		ctx = context.Background()
+	}
 
 	timer := time.NewTimer(duration)
 	defer timer.Stop()
@@ -856,6 +1197,8 @@ func waitForShutdownOrTimer(shutdown <-chan struct{}, duration time.Duration) bo
 	select {
 	case <-shutdown:
 		return false
+	case <-ctx.Done():
+		return false
 	case <-timer.C:
 		return true
 	}

weed/admin/plugin/plugin_scheduler_test.go

@@ -1,6 +1,7 @@
 package plugin
 
 import (
+	"context"
 	"fmt"
 	"testing"
 	"time"
@@ -28,6 +29,7 @@ func TestLoadSchedulerPolicyUsesAdminConfig(t *testing.T) {
 			PerWorkerExecutionConcurrency: 2,
 			RetryLimit:                    4,
 			RetryBackoffSeconds:           7,
+			JobTypeMaxRuntimeSeconds:      1800,
 		},
 	})
 	if err != nil {
@@ -53,6 +55,9 @@ func TestLoadSchedulerPolicyUsesAdminConfig(t *testing.T) {
 	if policy.RetryLimit != 4 {
 		t.Fatalf("unexpected retry limit: got=%d", policy.RetryLimit)
 	}
+	if policy.JobTypeMaxRuntime != 30*time.Minute {
+		t.Fatalf("unexpected max runtime: got=%v", policy.JobTypeMaxRuntime)
+	}
 }
 
 func TestLoadSchedulerPolicyUsesDescriptorDefaultsWhenConfigMissing(t *testing.T) {
@@ -75,6 +80,7 @@ func TestLoadSchedulerPolicyUsesDescriptorDefaultsWhenConfigMissing(t *testing.T
 			PerWorkerExecutionConcurrency: 2,
 			RetryLimit:                    3,
 			RetryBackoffSeconds:           6,
+			JobTypeMaxRuntimeSeconds:      1200,
 		},
 	})
 	if err != nil {
@@ -97,6 +103,9 @@ func TestLoadSchedulerPolicyUsesDescriptorDefaultsWhenConfigMissing(t *testing.T
 	if policy.PerWorkerConcurrency != 2 {
 		t.Fatalf("unexpected per-worker concurrency: got=%d", policy.PerWorkerConcurrency)
 	}
+	if policy.JobTypeMaxRuntime != 20*time.Minute {
+		t.Fatalf("unexpected max runtime: got=%v", policy.JobTypeMaxRuntime)
+	}
 }
 
 func TestReserveScheduledExecutorRespectsPerWorkerLimit(t *testing.T) {
@@ -126,13 +135,13 @@ func TestReserveScheduledExecutorRespectsPerWorkerLimit(t *testing.T) {
 		ExecutorReserveBackoff: time.Millisecond,
 	}
 
-	executor1, release1, err := pluginSvc.reserveScheduledExecutor("balance", policy)
+	executor1, release1, err := pluginSvc.reserveScheduledExecutor(context.Background(), "balance", policy)
 	if err != nil {
 		t.Fatalf("reserve executor 1: %v", err)
 	}
 	defer release1()
 
-	executor2, release2, err := pluginSvc.reserveScheduledExecutor("balance", policy)
+	executor2, release2, err := pluginSvc.reserveScheduledExecutor(context.Background(), "balance", policy)
 	if err != nil {
 		t.Fatalf("reserve executor 2: %v", err)
 	}
@@ -254,7 +263,7 @@ func TestReserveScheduledExecutorTimesOutWhenNoExecutor(t *testing.T) {
 
 	start := time.Now()
 	pluginSvc.Shutdown()
-	_, _, err = pluginSvc.reserveScheduledExecutor("missing-job-type", policy)
+	_, _, err = pluginSvc.reserveScheduledExecutor(context.Background(), "missing-job-type", policy)
 	if err == nil {
 		t.Fatalf("expected reservation shutdown error")
 	}
@@ -285,7 +294,7 @@ func TestReserveScheduledExecutorWaitsForWorkerCapacity(t *testing.T) {
 		ExecutorReserveBackoff: 5 * time.Millisecond,
 	}
 
-	_, release1, err := pluginSvc.reserveScheduledExecutor("balance", policy)
+	_, release1, err := pluginSvc.reserveScheduledExecutor(context.Background(), "balance", policy)
 	if err != nil {
 		t.Fatalf("reserve executor 1: %v", err)
 	}
@@ -296,7 +305,7 @@ func TestReserveScheduledExecutorWaitsForWorkerCapacity(t *testing.T) {
 	}
 	secondReserveCh := make(chan reserveResult, 1)
 	go func() {
-		_, release2, reserveErr := pluginSvc.reserveScheduledExecutor("balance", policy)
+		_, release2, reserveErr := pluginSvc.reserveScheduledExecutor(context.Background(), "balance", policy)
 		if release2 != nil {
 			release2()
 		}
@@ -394,6 +403,7 @@ func TestListSchedulerStatesIncludesPolicyAndState(t *testing.T) {
 			PerWorkerExecutionConcurrency: 2,
 			RetryLimit:                    1,
 			RetryBackoffSeconds:           9,
+			JobTypeMaxRuntimeSeconds:      900,
 		},
 	})
 	if err != nil {
@@ -446,6 +456,9 @@ func TestListSchedulerStatesIncludesPolicyAndState(t *testing.T) {
 	if state.ExecutionTimeoutSeconds != 90 {
 		t.Fatalf("unexpected execution timeout: got=%d", state.ExecutionTimeoutSeconds)
 	}
+	if state.JobTypeMaxRuntimeSeconds != 900 {
+		t.Fatalf("unexpected job type max runtime: got=%d", state.JobTypeMaxRuntimeSeconds)
+	}
 	if state.MaxJobsPerDetection != 80 {
 		t.Fatalf("unexpected max jobs per detection: got=%d", state.MaxJobsPerDetection)
 	}

weed/admin/plugin/scheduler_config.go

@@ -0,0 +1,5 @@
+package plugin
+
+import "time"
+
+const defaultSchedulerIdleSleep = 61 * time.Second

weed/admin/plugin/scheduler_status.go

@@ -9,6 +9,12 @@ import (
 type SchedulerStatus struct {
 	Now                  time.Time                `json:"now"`
 	SchedulerTickSeconds int                      `json:"scheduler_tick_seconds"`
+	IdleSleepSeconds     int                      `json:"idle_sleep_seconds,omitempty"`
+	NextDetectionAt      *time.Time               `json:"next_detection_at,omitempty"`
+	CurrentJobType       string                   `json:"current_job_type,omitempty"`
+	CurrentPhase         string                   `json:"current_phase,omitempty"`
+	LastIterationHadJobs bool                     `json:"last_iteration_had_jobs,omitempty"`
+	LastIterationDoneAt  *time.Time               `json:"last_iteration_done_at,omitempty"`
 	Waiting              []SchedulerWaitingStatus `json:"waiting,omitempty"`
 	InProcessJobs        []SchedulerJobStatus     `json:"in_process_jobs,omitempty"`
 	JobTypes             []SchedulerJobTypeStatus `json:"job_types,omitempty"`
@@ -56,6 +62,19 @@ type schedulerDetectionInfo struct {
 	lastSkippedReason string
 }
 
+type schedulerRunInfo struct {
+	lastRunStartedAt   time.Time
+	lastRunCompletedAt time.Time
+	lastRunStatus      string
+}
+
+type schedulerLoopState struct {
+	currentJobType         string
+	currentPhase           string
+	lastIterationHadJobs   bool
+	lastIterationCompleted time.Time
+}
+
 func (r *Plugin) recordSchedulerDetectionSuccess(jobType string, count int) {
 	if r == nil {
 		return
@@ -122,12 +141,101 @@ func (r *Plugin) snapshotSchedulerDetection(jobType string) schedulerDetectionIn
 	return *info
 }
 
+func (r *Plugin) recordSchedulerRunStart(jobType string) {
+	if r == nil {
+		return
+	}
+	r.schedulerRunMu.Lock()
+	defer r.schedulerRunMu.Unlock()
+	info := r.schedulerRun[jobType]
+	if info == nil {
+		info = &schedulerRunInfo{}
+		r.schedulerRun[jobType] = info
+	}
+	info.lastRunStartedAt = time.Now().UTC()
+	info.lastRunStatus = ""
+}
+
+func (r *Plugin) recordSchedulerRunComplete(jobType, status string) {
+	if r == nil {
+		return
+	}
+	r.schedulerRunMu.Lock()
+	defer r.schedulerRunMu.Unlock()
+	info := r.schedulerRun[jobType]
+	if info == nil {
+		info = &schedulerRunInfo{}
+		r.schedulerRun[jobType] = info
+	}
+	info.lastRunCompletedAt = time.Now().UTC()
+	info.lastRunStatus = status
+}
+
+func (r *Plugin) snapshotSchedulerRun(jobType string) schedulerRunInfo {
+	if r == nil {
+		return schedulerRunInfo{}
+	}
+	r.schedulerRunMu.Lock()
+	defer r.schedulerRunMu.Unlock()
+	info := r.schedulerRun[jobType]
+	if info == nil {
+		return schedulerRunInfo{}
+	}
+	return *info
+}
+
+func (r *Plugin) setSchedulerLoopState(jobType, phase string) {
+	if r == nil {
+		return
+	}
+	r.schedulerLoopMu.Lock()
+	r.schedulerLoopState.currentJobType = jobType
+	r.schedulerLoopState.currentPhase = phase
+	r.schedulerLoopMu.Unlock()
+}
+
+func (r *Plugin) recordSchedulerIterationComplete(hadJobs bool) {
+	if r == nil {
+		return
+	}
+	r.schedulerLoopMu.Lock()
+	r.schedulerLoopState.lastIterationHadJobs = hadJobs
+	r.schedulerLoopState.lastIterationCompleted = time.Now().UTC()
+	r.schedulerLoopMu.Unlock()
+}
+
+func (r *Plugin) snapshotSchedulerLoopState() schedulerLoopState {
+	if r == nil {
+		return schedulerLoopState{}
+	}
+	r.schedulerLoopMu.Lock()
+	defer r.schedulerLoopMu.Unlock()
+	return r.schedulerLoopState
+}
+
 func (r *Plugin) GetSchedulerStatus() SchedulerStatus {
 	now := time.Now().UTC()
+	loopState := r.snapshotSchedulerLoopState()
 	status := SchedulerStatus{
 		Now:                  now,
 		SchedulerTickSeconds: int(secondsFromDuration(r.schedulerTick)),
 		InProcessJobs:        r.listInProcessJobs(now),
+		IdleSleepSeconds:     int(defaultSchedulerIdleSleep / time.Second),
+		CurrentJobType:       loopState.currentJobType,
+		CurrentPhase:         loopState.currentPhase,
+		LastIterationHadJobs: loopState.lastIterationHadJobs,
+	}
+	nextDetectionAt := r.earliestNextDetectionAt()
+	if nextDetectionAt.IsZero() && loopState.currentPhase == "sleeping" && !loopState.lastIterationCompleted.IsZero() {
+		nextDetectionAt = loopState.lastIterationCompleted.Add(defaultSchedulerIdleSleep)
+	}
+	if !nextDetectionAt.IsZero() {
+		at := nextDetectionAt
+		status.NextDetectionAt = &at
+	}
+	if !loopState.lastIterationCompleted.IsZero() {
+		at := loopState.lastIterationCompleted
+		status.LastIterationDoneAt = &at
 	}
 
 	states, err := r.ListSchedulerStates()

weed/admin/plugin/types.go

@@ -90,6 +90,7 @@ type SchedulerJobTypeState struct {
 	DetectionIntervalSeconds      int32      `json:"detection_interval_seconds,omitempty"`
 	DetectionTimeoutSeconds       int32      `json:"detection_timeout_seconds,omitempty"`
 	ExecutionTimeoutSeconds       int32      `json:"execution_timeout_seconds,omitempty"`
+	JobTypeMaxRuntimeSeconds      int32      `json:"job_type_max_runtime_seconds,omitempty"`
 	MaxJobsPerDetection           int32      `json:"max_jobs_per_detection,omitempty"`
 	GlobalExecutionConcurrency    int        `json:"global_execution_concurrency,omitempty"`
 	PerWorkerExecutionConcurrency int        `json:"per_worker_execution_concurrency,omitempty"`
@@ -98,6 +99,9 @@ type SchedulerJobTypeState struct {
 	DetectorAvailable             bool       `json:"detector_available"`
 	DetectorWorkerID              string     `json:"detector_worker_id,omitempty"`
 	ExecutorWorkerCount           int        `json:"executor_worker_count"`
+	LastRunStartedAt              *time.Time `json:"last_run_started_at,omitempty"`
+	LastRunCompletedAt            *time.Time `json:"last_run_completed_at,omitempty"`
+	LastRunStatus                 string     `json:"last_run_status,omitempty"`
 }
 
 func timeToPtr(t time.Time) *time.Time {

weed/admin/static/css/admin.css

@@ -1,5 +1,9 @@
 /* SeaweedFS Dashboard Custom Styles */
 
+:root {
+    --navbar-height: 56px;
+}
+
 /* Link colors - muted */
 a {
     color: #5b7c99;
@@ -12,7 +16,7 @@ a:hover {
 /* Sidebar Styles */
 .sidebar {
     position: fixed;
-    top: 56px;
+    top: var(--navbar-height);
     bottom: 0;
     left: 0;
     z-index: 100;
@@ -51,13 +55,32 @@ main {
 
 @media (max-width: 767.98px) {
     .sidebar {
-        top: 5rem;
+        top: var(--navbar-height);
+        padding-top: 0;
+        width: 240px;
+        background-color: #f8f9fa !important;
+        z-index: 1050;
+    }
+    .sidebar.show ~ .sidebar-backdrop {
+        display: block;
     }
     main {
         margin-left: 0;
     }
 }
 
+/* Sidebar backdrop for mobile overlay */
+.sidebar-backdrop {
+    display: none;
+    position: fixed;
+    top: var(--navbar-height);
+    left: 0;
+    right: 0;
+    bottom: 0;
+    background-color: rgba(0, 0, 0, 0.5);
+    z-index: 1040;
+}
+
 /* Custom card styles */
 .border-left-primary {
     border-left: 0.25rem solid #6b8caf !important;
@@ -262,6 +285,11 @@ main {
     box-shadow: 0 0.15rem 1.75rem 0 rgba(58, 59, 69, 0.15) !important;
 }
 
+/* Navbar user icon color */
+.navbar-toggler .fa-user {
+    color: rgba(255, 255, 255, 0.75);
+}
+
 /* Collapsible menu styles */
 .nav-link[data-bs-toggle="collapse"] {
     position: relative;

weed/admin/static/js/admin.js

@@ -31,6 +31,9 @@ function initializeDashboard() {
 
     // Set up submenu behavior
     setupSubmenuBehavior();
+
+    // Set up mobile sidebar behavior
+    setupMobileSidebar();
 }
 
 // HTMX event listeners
@@ -194,6 +197,33 @@ function setupSubmenuBehavior() {
 
 }
 
+// Mobile sidebar toggle and backdrop behavior
+function setupMobileSidebar() {
+    const sidebar = document.getElementById('sidebarMenu');
+    const backdrop = document.getElementById('sidebarBackdrop');
+    if (!sidebar || !backdrop) return;
+
+    const hideSidebar = () => {
+        const bsCollapse = bootstrap.Collapse.getInstance(sidebar);
+        if (bsCollapse) {
+            bsCollapse.hide();
+        }
+    };
+
+    // Close sidebar when backdrop is clicked
+    backdrop.addEventListener('click', hideSidebar);
+
+    // Close sidebar when a nav link is clicked (on mobile)
+    const sidebarToggler = document.querySelector("button[data-bs-target='#sidebarMenu']");
+    sidebar.querySelectorAll('a.nav-link:not([data-bs-toggle="collapse"])').forEach(function (link) {
+        link.addEventListener('click', function () {
+            if (sidebarToggler && getComputedStyle(sidebarToggler).display !== 'none') {
+                hideSidebar();
+            }
+        });
+    });
+}
+
 // Loading indicator functions
 function showLoadingIndicator() {
     const indicator = document.getElementById('loading-indicator');
@@ -448,7 +478,7 @@ async function handleCreateBucket(event) {
 
         if (response.ok) {
             // Success
-            showAlert('success', `Bucket "${bucketData.name}" created successfully!`);
+            showAlert(`Bucket "${bucketData.name}" created successfully!`, 'success');
 
             // Close modal
             const modal = bootstrap.Modal.getInstance(document.getElementById('createBucketModal'));
@@ -463,11 +493,11 @@ async function handleCreateBucket(event) {
             }, 1500);
         } else {
             // Error
-            showAlert('danger', result.error || 'Failed to create bucket');
+            showAlert(result.error || 'Failed to create bucket', 'danger');
         }
     } catch (error) {
         console.error('Error creating bucket:', error);
-        showAlert('danger', 'Network error occurred while creating bucket');
+        showAlert('Network error occurred while creating bucket', 'danger');
     }
 }
 
@@ -508,7 +538,7 @@ async function deleteBucket() {
 
         if (response.ok) {
             // Success
-            showAlert('success', `Bucket "${bucketToDelete}" deleted successfully!`);
+            showAlert(`Bucket "${bucketToDelete}" deleted successfully!`, 'success');
 
             // Close modal
             const modal = bootstrap.Modal.getInstance(document.getElementById('deleteBucketModal'));
@@ -520,11 +550,11 @@ async function deleteBucket() {
             }, 1500);
         } else {
             // Error
-            showAlert('danger', result.error || 'Failed to delete bucket');
+            showAlert(result.error || 'Failed to delete bucket', 'danger');
         }
     } catch (error) {
         console.error('Error deleting bucket:', error);
-        showAlert('danger', 'Network error occurred while deleting bucket');
+        showAlert('Network error occurred while deleting bucket', 'danger');
     }
 
     bucketToDelete = '';
@@ -579,38 +609,7 @@ function exportBucketList() {
     window.URL.revokeObjectURL(url);
 }
 
-// Show alert message
-function showAlert(type, message) {
-    // Remove existing alerts
-    const existingAlerts = document.querySelectorAll('.alert-floating');
-    existingAlerts.forEach(alert => alert.remove());
-
-    // Create new alert
-    const alert = document.createElement('div');
-    alert.className = `alert alert-${type} alert-dismissible fade show alert-floating`;
-    alert.style.cssText = `
-        position: fixed;
-        top: 20px;
-        right: 20px;
-        z-index: 9999;
-        min-width: 300px;
-        box-shadow: 0 4px 6px rgba(0, 0, 0, 0.1);
-    `;
-
-    alert.innerHTML = `
-        ${message}
-        <button type="button" class="btn-close" data-bs-dismiss="alert" aria-label="Close"></button>
-    `;
-
-    document.body.appendChild(alert);
-
-    // Auto-remove after 5 seconds
-    setTimeout(() => {
-        if (alert.parentNode) {
-            alert.remove();
-        }
-    }, 5000);
-}
+// showAlert is provided by modal-alerts.js with signature: showAlert(message, type)
 
 // Format date for display
 function formatDate(date) {
@@ -621,7 +620,7 @@ function formatDate(date) {
 function adminCopyToClipboard(text) {
     if (navigator.clipboard && navigator.clipboard.writeText) {
         navigator.clipboard.writeText(text).then(() => {
-            showAlert('success', 'Copied to clipboard!');
+            showAlert('Copied to clipboard!', 'success');
         }).catch(err => {
             console.error('Failed to copy text: ', err);
             fallbackCopyText(text);
@@ -647,13 +646,13 @@ function fallbackCopyText(text) {
     try {
         const successful = document.execCommand('copy');
         if (successful) {
-            showAlert('success', 'Copied to clipboard!');
+            showAlert('Copied to clipboard!', 'success');
         } else {
-            showAlert('danger', 'Failed to copy to clipboard');
+            showAlert('Failed to copy to clipboard', 'danger');
         }
     } catch (err) {
         console.error('Fallback copy failed: ', err);
-        showAlert('danger', 'Failed to copy to clipboard');
+        showAlert('Failed to copy to clipboard', 'danger');
     }
 
     document.body.removeChild(textArea);
@@ -734,7 +733,7 @@ function exportVolumes() {
 function exportCollections() {
     const table = document.getElementById('collectionsTable');
     if (!table) {
-        showAlert('error', 'Collections table not found');
+        showAlert('Collections table not found', 'error');
         return;
     }
 
@@ -770,7 +769,7 @@ function exportCollections() {
 function exportMasters() {
     const table = document.getElementById('mastersTable');
     if (!table) {
-        showAlert('error', 'Masters table not found');
+        showAlert('Masters table not found', 'error');
         return;
     }
 
@@ -804,7 +803,7 @@ function exportMasters() {
 function exportFilers() {
     const table = document.getElementById('filersTable');
     if (!table) {
-        showAlert('error', 'Filers table not found');
+        showAlert('Filers table not found', 'error');
         return;
     }
 
@@ -840,7 +839,7 @@ function exportFilers() {
 function exportUsers() {
     const table = document.getElementById('usersTable');
     if (!table) {
-        showAlert('error', 'Users table not found');
+        showAlert('Users table not found', 'error');
         return;
     }
 
@@ -990,7 +989,7 @@ function confirmDeleteSelected() {
     const selectedPaths = getSelectedFilePaths();
 
     if (selectedPaths.length === 0) {
-        showAlert('warning', 'No files selected');
+        showAlert('No files selected', 'warning');
         return;
     }
 
@@ -1011,7 +1010,7 @@ function confirmDeleteSelected() {
 // Delete multiple selected files
 async function deleteSelectedFiles(filePaths) {
     if (!filePaths || filePaths.length === 0) {
-        showAlert('warning', 'No files selected');
+        showAlert('No files selected', 'warning');
         return;
     }
 
@@ -1035,9 +1034,9 @@ async function deleteSelectedFiles(filePaths) {
 
             if (result.deleted > 0) {
                 if (result.failed === 0) {
-                    showAlert('success', `Successfully deleted ${result.deleted} item(s)`);
+                    showAlert(`Successfully deleted ${result.deleted} item(s)`, 'success');
                 } else {
-                    showAlert('warning', `Deleted ${result.deleted} item(s), failed to delete ${result.failed} item(s)`);
+                    showAlert(`Deleted ${result.deleted} item(s), failed to delete ${result.failed} item(s)`, 'warning');
                     if (result.errors && result.errors.length > 0) {
                         console.warn('Deletion errors:', result.errors);
                     }
@@ -1052,15 +1051,15 @@ async function deleteSelectedFiles(filePaths) {
                 if (result.errors && result.errors.length > 0) {
                     errorMessage += ': ' + result.errors.join(', ');
                 }
-                showAlert('error', errorMessage);
+                showAlert(errorMessage, 'error');
             }
         } else {
             const error = await response.json();
-            showAlert('error', `Failed to delete files: ${error.error || 'Unknown error'}`);
+            showAlert(`Failed to delete files: ${error.error || 'Unknown error'}`, 'error');
         }
     } catch (error) {
         console.error('Delete error:', error);
-        showAlert('error', 'Failed to delete files');
+        showAlert('Failed to delete files', 'error');
     } finally {
         // Re-enable the button
         deleteBtn.disabled = false;
@@ -1281,7 +1280,7 @@ async function submitUploadFile() {
 function exportFileList() {
     const table = document.getElementById('fileTable');
     if (!table) {
-        showAlert('error', 'File table not found');
+        showAlert('File table not found', 'error');
         return;
     }
 
@@ -1327,7 +1326,7 @@ async function viewFile(filePath) {
 
         if (!response.ok) {
             const error = await response.json();
-            showAlert('error', `Failed to view file: ${error.error || 'Unknown error'}`);
+            showAlert(`Failed to view file: ${error.error || 'Unknown error'}`, 'error');
             return;
         }
 
@@ -1336,7 +1335,7 @@ async function viewFile(filePath) {
 
     } catch (error) {
         console.error('View file error:', error);
-        showAlert('error', 'Failed to view file');
+        showAlert('Failed to view file', 'error');
     }
 }
 
@@ -1347,7 +1346,7 @@ async function showProperties(filePath) {
 
         if (!response.ok) {
             const error = await response.json();
-            showAlert('error', `Failed to get file properties: ${error.error || 'Unknown error'}`);
+            showAlert(`Failed to get file properties: ${error.error || 'Unknown error'}`, 'error');
             return;
         }
 
@@ -1356,7 +1355,7 @@ async function showProperties(filePath) {
 
     } catch (error) {
         console.error('Properties error:', error);
-        showAlert('error', 'Failed to get file properties');
+        showAlert('Failed to get file properties', 'error');
     }
 }
 
@@ -1383,16 +1382,16 @@ async function deleteFile(filePath) {
         });
 
         if (response.ok) {
-            showAlert('success', `Successfully deleted "${filePath}"`);
+            showAlert(`Successfully deleted "${filePath}"`, 'success');
             // Reload the page to update the file list
             window.location.reload();
         } else {
             const error = await response.json();
-            showAlert('error', `Failed to delete file: ${error.error || 'Unknown error'}`);
+            showAlert(`Failed to delete file: ${error.error || 'Unknown error'}`, 'error');
         }
     } catch (error) {
         console.error('Delete error:', error);
-        showAlert('error', 'Failed to delete file');
+        showAlert('Failed to delete file', 'error');
     }
 }
 
@@ -1707,7 +1706,7 @@ async function handleUpdateQuota(event) {
 
         if (response.ok) {
             // Success
-            showAlert('success', `Quota for bucket "${bucketName}" updated successfully!`);
+            showAlert(`Quota for bucket "${bucketName}" updated successfully!`, 'success');
 
             // Close modal
             const modal = bootstrap.Modal.getInstance(document.getElementById('manageQuotaModal'));
@@ -1719,11 +1718,11 @@ async function handleUpdateQuota(event) {
             }, 1500);
         } else {
             // Error
-            showAlert('danger', result.error || 'Failed to update bucket quota');
+            showAlert(result.error || 'Failed to update bucket quota', 'danger');
         }
     } catch (error) {
         console.error('Error updating bucket quota:', error);
-        showAlert('danger', 'Network error occurred while updating bucket quota');
+        showAlert('Network error occurred while updating bucket quota', 'danger');
     }
 }
 
@@ -2197,10 +2196,6 @@ function showNewAccessKeyModal(accessKeyData) {
             <i class="fas fa-check-circle me-2"></i>
             <strong>Success!</strong> Your new access key has been created.
         </div>
-        <div class="alert alert-warning">
-            <i class="fas fa-exclamation-triangle me-2"></i>
-            <strong>Important:</strong> This is the only time the secret key will be displayed. Please save it securely.
-        </div>
         <div class="mb-3">
             <label class="form-label"><strong>Access Key:</strong></label>
             <div class="input-group">
@@ -2244,21 +2239,21 @@ function copyFromInput(inputId) {
         try {
             const successful = document.execCommand('copy');
             if (successful) {
-                showAlert('success', 'Copied to clipboard!');
+                showAlert('Copied to clipboard!', 'success');
             } else {
                 // Try modern clipboard API as fallback
                 navigator.clipboard.writeText(input.value).then(() => {
-                    showAlert('success', 'Copied to clipboard!');
+                    showAlert('Copied to clipboard!', 'success');
                 }).catch(() => {
-                    showAlert('danger', 'Failed to copy');
+                    showAlert('Failed to copy', 'danger');
                 });
             }
         } catch (err) {
             // Try modern clipboard API as fallback
             navigator.clipboard.writeText(input.value).then(() => {
-                showAlert('success', 'Copied to clipboard!');
+                showAlert('Copied to clipboard!', 'success');
             }).catch(() => {
-                showAlert('danger', 'Failed to copy');
+                showAlert('Failed to copy', 'danger');
             });
         }
     }

weed/admin/static/js/iam-utils.js

@@ -25,6 +25,29 @@ async function deleteUser(username) {
     }, 'Are you sure you want to delete this user? This action cannot be undone.');
 }
 
+// Delete group function
+async function deleteGroup(name) {
+    showDeleteConfirm(name, async function () {
+        try {
+            const encodedName = encodeURIComponent(name);
+            const response = await fetch(`/api/groups/${encodedName}`, {
+                method: 'DELETE'
+            });
+
+            if (response.ok) {
+                showAlert('Group deleted successfully', 'success');
+                setTimeout(() => window.location.reload(), 1000);
+            } else {
+                const error = await response.json().catch(() => ({}));
+                showAlert('Failed to delete group: ' + (error.error || 'Unknown error'), 'error');
+            }
+        } catch (error) {
+            console.error('Error deleting group:', error);
+            showAlert('Failed to delete group: ' + error.message, 'error');
+        }
+    }, 'Are you sure you want to delete this group? This action cannot be undone.');
+}
+
 // Delete access key function
 async function deleteAccessKey(username, accessKey) {
     showDeleteConfirm(accessKey, async function () {

weed/admin/topology/task_management.go

@@ -17,20 +17,30 @@ func (at *ActiveTopology) AssignTask(taskID string) error {
 		return fmt.Errorf("pending task %s not found", taskID)
 	}
 
-	// Check if all destination disks have sufficient capacity to reserve
-	for _, dest := range task.Destinations {
-		targetKey := fmt.Sprintf("%s:%d", dest.TargetServer, dest.TargetDisk)
-		if targetDisk, exists := at.disks[targetKey]; exists {
-			availableCapacity := at.getEffectiveAvailableCapacityUnsafe(targetDisk)
+	// Skip capacity check if topology hasn't been populated yet
+	if len(at.disks) == 0 {
+		glog.Warningf("AssignTask %s: topology has no disks yet, skipping capacity check", taskID)
+	} else {
+		// Check if all destination disks have sufficient capacity to reserve
+		for _, dest := range task.Destinations {
+			targetKey := fmt.Sprintf("%s:%d", dest.TargetServer, dest.TargetDisk)
+			if targetDisk, exists := at.disks[targetKey]; exists {
+				availableCapacity := at.getEffectiveAvailableCapacityUnsafe(targetDisk)
 
-			// Check if we have enough total capacity using the improved unified comparison
-			if !availableCapacity.CanAccommodate(dest.StorageChange) {
-				return fmt.Errorf("insufficient capacity on target disk %s:%d. Available: %+v, Required: %+v",
-					dest.TargetServer, dest.TargetDisk, availableCapacity, dest.StorageChange)
+				// Check if we have enough total capacity using the improved unified comparison
+				if !availableCapacity.CanAccommodate(dest.StorageChange) {
+					return fmt.Errorf("insufficient capacity on target disk %s:%d. Available: %+v, Required: %+v",
+						dest.TargetServer, dest.TargetDisk, availableCapacity, dest.StorageChange)
+				}
+			} else if dest.TargetServer != "" {
+				// Fail fast if destination disk is not found in topology
+				var existingKeys []string
+				for k := range at.disks {
+					existingKeys = append(existingKeys, k)
+				}
+				glog.Warningf("destination disk %s not found in topology. Existing disk keys: %v", targetKey, existingKeys)
+				return fmt.Errorf("destination disk %s not found in topology", targetKey)
 			}
-		} else if dest.TargetServer != "" {
-			// Fail fast if destination disk is not found in topology
-			return fmt.Errorf("destination disk %s not found in topology", targetKey)
 		}
 	}
 
@@ -262,6 +272,40 @@ func (at *ActiveTopology) HasAnyTask(volumeID uint32) bool {
 	return at.HasTask(volumeID, TaskTypeNone)
 }
 
+// GetTaskServerAdjustments returns per-server volume count adjustments for
+// pending and assigned tasks of the given type. For each task, source servers
+// are decremented and destination servers are incremented, reflecting the
+// projected volume distribution once in-flight tasks complete.
+func (at *ActiveTopology) GetTaskServerAdjustments(taskType TaskType) map[string]int {
+	at.mutex.RLock()
+	defer at.mutex.RUnlock()
+
+	adjustments := make(map[string]int)
+	for _, task := range at.pendingTasks {
+		if task.TaskType != taskType {
+			continue
+		}
+		for _, src := range task.Sources {
+			adjustments[src.SourceServer]--
+		}
+		for _, dst := range task.Destinations {
+			adjustments[dst.TargetServer]++
+		}
+	}
+	for _, task := range at.assignedTasks {
+		if task.TaskType != taskType {
+			continue
+		}
+		for _, src := range task.Sources {
+			adjustments[src.SourceServer]--
+		}
+		for _, dst := range task.Destinations {
+			adjustments[dst.TargetServer]++
+		}
+	}
+	return adjustments
+}
+
 // calculateSourceStorageImpact calculates storage impact for sources based on task type and cleanup type
 func (at *ActiveTopology) calculateSourceStorageImpact(taskType TaskType, cleanupType SourceCleanupType, volumeSize int64) StorageSlotChange {
 	switch taskType {