Files
seaweedfs/weed/server/master_grpc_server_assign.go
T
Chris LuandGitHub ab7be7867d security: hot-reload JWT signing keys on SIGHUP (#9826)
* security: reload JWT signing keys on SIGHUP

Signing keys were read once in the server constructors and never
refreshed. After a key rotation (Secret update, divergent reads) the
in-memory key stayed stale and every request kept failing "wrong jwt"
until the affected process was restarted.

Add Guard.UpdateSigningKeys and call it from the master, volume and
filer reload paths and the s3 reload hook, next to the existing
whitelist refresh. Make the global chunk-read JWT cache reloadable via
an atomic swap, and register the master's Reload with grace.OnReload --
it was never wired, so the master ignored SIGHUP entirely.

Mirror the same refresh in the Rust volume server's SIGHUP handler.

* security: swap signing keys behind an atomic pointer

Addresses review feedback on the in-place key swap: SigningKey is a
[]byte, so reassigning the Guard fields while a request handler reads
them is a data race that can tear the multi-word slice header and read
out of bounds.

Hold the four signing-key fields in an immutable signingConfig snapshot
behind atomic.Pointer; UpdateSigningKeys swaps the whole pointer, so a
reader sees either the old keys or the new ones. Reads go through new
SigningKey/ExpiresAfterSec/ReadSigningKey/ReadExpiresAfterSec accessors.

The Rust guard is already safe: every read and the SIGHUP write go
through the shared RwLock<Guard>.

* security: fold whitelist + auth state into the atomic snapshot

Review follow-up. UpdateSigningKeys still wrote isWriteActive while the
request path read it (and the whitelist maps) unsynchronized, so a SIGHUP
under load could expose an inconsistent mix of activation bits and
whitelist contents.

Move all hot-reloadable Guard state -- keys, expirations, whitelist, and
the activation flags -- into a single immutable guardState swapped behind
one atomic.Pointer. The Update* methods take a small mutex to serialize
the read-modify-write; readers stay lock-free. The concurrency test now
also rotates the whitelist and probes IsWhiteListed under -race.

Also read each signing key once per branch in the volume/filer JWT auth
checks, so a reload landing mid-check can't take the allow-fast-path
after auth was enabled or verify against a different key than the branch
saw.
2026-06-04 22:26:08 -07:00

167 lines
4.9 KiB
Go

package weed_server
import (
"context"
"fmt"
"strings"
"time"
"github.com/seaweedfs/seaweedfs/weed/glog"
"github.com/seaweedfs/seaweedfs/weed/stats"
"github.com/seaweedfs/raft"
"github.com/seaweedfs/seaweedfs/weed/pb/master_pb"
"github.com/seaweedfs/seaweedfs/weed/security"
"github.com/seaweedfs/seaweedfs/weed/storage/needle"
"github.com/seaweedfs/seaweedfs/weed/storage/super_block"
"github.com/seaweedfs/seaweedfs/weed/storage/types"
"github.com/seaweedfs/seaweedfs/weed/topology"
"google.golang.org/grpc/codes"
"google.golang.org/grpc/status"
)
func (ms *MasterServer) StreamAssign(server master_pb.Seaweed_StreamAssignServer) error {
for {
req, err := server.Recv()
if err != nil {
glog.Errorf("StreamAssign failed to receive: %v", err)
return err
}
resp, err := ms.Assign(context.Background(), req)
if err != nil {
// Return transient errors (e.g. warmup) as in-band error responses
// instead of killing the stream, so pooled connections survive.
if st, ok := status.FromError(err); ok && st.Code() == codes.Unavailable {
glog.V(1).Infof("StreamAssign transient error: %v", err)
resp = &master_pb.AssignResponse{Error: st.Message()}
} else {
glog.Errorf("StreamAssign failed to assign: %v", err)
return err
}
}
if err = server.Send(resp); err != nil {
glog.Errorf("StreamAssign failed to send: %v", err)
return err
}
}
}
func (ms *MasterServer) Assign(ctx context.Context, req *master_pb.AssignRequest) (*master_pb.AssignResponse, error) {
if !ms.Topo.IsLeader() {
return nil, raft.NotLeaderError
}
if req.Count == 0 {
req.Count = 1
}
if req.Replication == "" {
req.Replication = ms.option.DefaultReplicaPlacement
}
replicaPlacement, err := super_block.NewReplicaPlacementFromString(req.Replication)
if err != nil {
return nil, err
}
ttl, err := needle.ReadTTL(req.Ttl)
if err != nil {
return nil, err
}
if ms.Topo.IsWarmingUp() {
return nil, status.Errorf(codes.Unavailable, "master is warming up, topology is still loading")
}
diskType := types.ToDiskType(req.DiskType)
ver := needle.GetCurrentVersion()
option := &topology.VolumeGrowOption{
Collection: req.Collection,
ReplicaPlacement: replicaPlacement,
Ttl: ttl,
DiskType: diskType,
Preallocate: ms.preallocateSize,
DataCenter: req.DataCenter,
Rack: req.Rack,
DataNode: req.DataNode,
MemoryMapMaxSizeMb: req.MemoryMapMaxSizeMb,
Version: uint32(ver),
}
if !ms.Topo.DataCenterExists(option.DataCenter) {
return nil, fmt.Errorf("data center %v not found in topology", option.DataCenter)
}
vl := ms.Topo.GetVolumeLayout(option.Collection, option.ReplicaPlacement, option.Ttl, option.DiskType)
if req.DiskType == "" {
if writable, _ := vl.GetWritableVolumeCount(); writable == 0 {
if hddVl := ms.Topo.GetVolumeLayout(option.Collection, option.ReplicaPlacement, option.Ttl, types.ToDiskType(types.HddType)); hddVl != nil {
if writable, _ := hddVl.GetWritableVolumeCount(); writable > 0 {
option.DiskType = types.ToDiskType(types.HddType)
vl = hddVl
}
}
}
}
vl.SetLastGrowCount(req.WritableVolumeCount)
var (
lastErr error
maxTimeout = time.Second * 10
startTime = time.Now()
)
for time.Now().Sub(startTime) < maxTimeout {
fid, count, dnList, shouldGrow, err := ms.Topo.PickForWrite(req.Count, option, vl, req.ExpectedDataSize)
if shouldGrow && !vl.HasGrowRequest() && !ms.option.VolumeGrowthDisabled {
if err != nil && ms.Topo.AvailableSpaceFor(option) <= 0 {
err = fmt.Errorf("%s and no free volumes left for %s", err.Error(), option.String())
}
vl.AddGrowRequest()
ms.volumeGrowthRequestChan <- &topology.VolumeGrowRequest{
Option: option,
Count: req.WritableVolumeCount,
Reason: "grpc assign",
}
}
if err != nil {
glog.V(1).Infof("assign %v %v: %v", req, option.String(), err)
stats.MasterPickForWriteErrorCounter.Inc()
lastErr = err
if (req.DataCenter != "" || req.Rack != "") && strings.Contains(err.Error(), topology.NoWritableVolumes) {
break
}
time.Sleep(200 * time.Millisecond)
continue
}
dn := dnList.Head()
if dn == nil {
continue
}
var replicas []*master_pb.Location
for _, r := range dnList.Rest() {
replicas = append(replicas, &master_pb.Location{
Url: r.Url(),
PublicUrl: r.PublicUrl,
GrpcPort: uint32(r.GrpcPort),
DataCenter: r.GetDataCenterId(),
})
}
return &master_pb.AssignResponse{
Fid: fid,
Location: &master_pb.Location{
Url: dn.Url(),
PublicUrl: dn.PublicUrl,
GrpcPort: uint32(dn.GrpcPort),
DataCenter: dn.GetDataCenterId(),
},
Count: count,
Auth: string(security.GenJwtForVolumeServer(ms.guard.SigningKey(), ms.guard.ExpiresAfterSec(), fid)),
Replicas: replicas,
}, nil
}
if lastErr != nil {
glog.V(0).Infof("assign %v %v: %v", req, option.String(), lastErr)
}
return nil, lastErr
}