From 2ac5aa72c7bdbda4935f330473886e3543a8923c Mon Sep 17 00:00:00 2001
From: Chris Lu <chrislusf@users.noreply.github.com>
Date: Wed, 10 Jun 2026 12:10:49 -0700
Subject: [PATCH] add elastic8 filer store for Elasticsearch 8 (#9916)

* elastic: fix listing against a missing or empty directory index

The refresh 404 leaked into the named return, so the first listing of a
directory whose index does not exist yet returned an error instead of an
empty result. Sorting also fails on an index with no documents
("No mapping found for [_id] in order to sort on"); unmapped_type
keeps the resumed-listing path working there.

* add elastic8 filer store for Elasticsearch 8

Elasticsearch 8 disables _id fielddata by default, so the elastic7
store's directory listings fail with "Fielddata access on the _id
field is disallowed". elastic8 uses the same client and configuration
options, but also indexes the document id as an Id field and sorts
listings on Id.keyword.
---
 weed/command/scaffold/filer.toml       | 15 ++++++++++
 weed/filer/elastic/v7/elastic_store.go | 39 +++++++++++++++++++++++---
 2 files changed, 50 insertions(+), 4 deletions(-)

diff --git a/weed/command/scaffold/filer.toml b/weed/command/scaffold/filer.toml
index 073acba83..e1aea5a72 100644
--- a/weed/command/scaffold/filer.toml
+++ b/weed/command/scaffold/filer.toml
@@ -366,6 +366,21 @@ healthcheck_enabled = false
 # increase the value is recommend, be sure the value in Elastic is greater or equal here
 index.max_result_window = 10000
 
+# for Elasticsearch 8.x clusters
+[elastic8]
+enabled = false
+servers = [
+    "http://localhost1:9200",
+    "http://localhost2:9200",
+    "http://localhost3:9200",
+]
+username = ""
+password = ""
+sniff_enabled = false
+healthcheck_enabled = false
+# increase the value is recommend, be sure the value in Elastic is greater or equal here
+index.max_result_window = 10000
+
 
 [arangodb] # in development dont use it
 enabled = false
diff --git a/weed/filer/elastic/v7/elastic_store.go b/weed/filer/elastic/v7/elastic_store.go
index db67b405d..3b3839bbc 100644
--- a/weed/filer/elastic/v7/elastic_store.go
+++ b/weed/filer/elastic/v7/elastic_store.go
@@ -34,6 +34,7 @@ var (
 
 type ESEntry struct {
 	ParentId string `json:"ParentId"`
+	Id       string `json:"Id,omitempty"`
 	Entry    *filer.Entry
 }
 
@@ -43,17 +44,33 @@ type ESKVEntry struct {
 
 func init() {
 	filer.Stores = append(filer.Stores, &ElasticStore{})
+	filer.Stores = append(filer.Stores, &Elastic8Store{})
 }
 
 type ElasticStore struct {
 	client      *elastic.Client
 	maxPageSize int
+	es8         bool
 }
 
 func (store *ElasticStore) GetName() string {
 	return "elastic7"
 }
 
+// Elastic8Store sorts listings on an indexed Id field since Elasticsearch 8 disallows _id fielddata.
+type Elastic8Store struct {
+	ElasticStore
+}
+
+func (store *Elastic8Store) GetName() string {
+	return "elastic8"
+}
+
+func (store *Elastic8Store) Initialize(configuration weed_util.Configuration, prefix string) (err error) {
+	store.es8 = true
+	return store.ElasticStore.Initialize(configuration, prefix)
+}
+
 func (store *ElasticStore) Initialize(configuration weed_util.Configuration, prefix string) (err error) {
 	options := []elastic.ClientOptionFunc{}
 	servers := configuration.GetStringSlice(prefix + "servers")
@@ -110,6 +127,9 @@ func (store *ElasticStore) InsertEntry(ctx context.Context, entry *filer.Entry)
 		ParentId: weed_util.Md5String([]byte(dir)),
 		Entry:    entry,
 	}
+	if store.es8 {
+		esEntry.Id = id
+	}
 	value, err := jsoniter.Marshal(esEntry)
 	if err != nil {
 		glog.ErrorfCtx(ctx, "insert entry(%s) %v.", string(entry.FullPath), err)
@@ -219,8 +239,10 @@ func (store *ElasticStore) listDirectoryEntries(
 	parentId := weed_util.Md5String([]byte(fullpath))
 	if _, err = store.client.Refresh(index).Do(ctx); err != nil {
 		if elastic.IsNotFound(err) {
-			store.client.CreateIndex(index).Do(ctx)
-			return
+			if _, err := store.client.CreateIndex(index).Do(ctx); err != nil {
+				return lastFileName, fmt.Errorf("create index(%s) %v", index, err)
+			}
+			return lastFileName, nil
 		}
 	}
 	for {
@@ -278,6 +300,15 @@ func (store *ElasticStore) listDirectoryEntries(
 	return
 }
 
+func (store *ElasticStore) listSorter() elastic.Sorter {
+	field := "_id"
+	if store.es8 {
+		field = "Id.keyword"
+	}
+	// unmapped_type tolerates indexes with no documents yet
+	return elastic.NewFieldSort(field).Desc().UnmappedType("keyword")
+}
+
 func (store *ElasticStore) search(ctx context.Context, index, parentId string) (result *elastic.SearchResult, err error) {
 	if count, err := store.client.Count(index).Do(ctx); err == nil && count == 0 {
 		return &elastic.SearchResult{
@@ -289,7 +320,7 @@ func (store *ElasticStore) search(ctx context.Context, index, parentId string) (
 		Index(index).
 		Query(elastic.NewMatchQuery("ParentId", parentId)).
 		Size(store.maxPageSize).
-		Sort("_id", false).
+		SortBy(store.listSorter()).
 		Do(ctx)
 	return queryResult, err
 }
@@ -300,7 +331,7 @@ func (store *ElasticStore) searchAfter(ctx context.Context, index, parentId, aft
 		Query(elastic.NewMatchQuery("ParentId", parentId)).
 		SearchAfter(after).
 		Size(store.maxPageSize).
-		Sort("_id", false).
+		SortBy(store.listSorter()).
 		Do(ctx)
 	return queryResult, err