diff options
| author | chrislusf <chris.lu@gmail.com> | 2025-12-03 20:00:39 -0800 |
|---|---|---|
| committer | Chris Lu <chrislusf@users.noreply.github.com> | 2025-12-03 20:52:27 -0800 |
| commit | fe60f2bf160c1191eb81765af2b2fc32a6d311b5 (patch) | |
| tree | 7b66a9faba9fdc14b2243d4e59b18000938ee104 /pkg/driver/nodeserver.go | |
| parent | 639573bdc0108da9c4d6fce7ffcb897ea1b71e14 (diff) | |
| download | seaweedfs-csi-driver-fe60f2bf160c1191eb81765af2b2fc32a6d311b5.tar.xz seaweedfs-csi-driver-fe60f2bf160c1191eb81765af2b2fc32a6d311b5.zip | |
fix: cleanup volume mutex on self-healing failure in NodePublishVolume
Address gemini-code-assist review - when cleanup or re-staging fails
during self-healing in NodePublishVolume, remove the volume mutex to
avoid leaving stale entries. This maintains consistency with
NodeStageVolume's error handling behavior.
Diffstat (limited to 'pkg/driver/nodeserver.go')
| -rw-r--r-- | pkg/driver/nodeserver.go | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/pkg/driver/nodeserver.go b/pkg/driver/nodeserver.go index e65038e..21e68bf 100644 --- a/pkg/driver/nodeserver.go +++ b/pkg/driver/nodeserver.go @@ -150,6 +150,7 @@ func (ns *NodeServer) NodePublishVolume(ctx context.Context, req *csi.NodePublis // Clean up stale staging path if it exists if err := cleanupStaleStagingPath(stagingTargetPath); err != nil { + ns.removeVolumeMutex(volumeID) return nil, status.Errorf(codes.Internal, "failed to cleanup stale staging path %s: %v", stagingTargetPath, err) } @@ -159,6 +160,7 @@ func (ns *NodeServer) NodePublishVolume(ctx context.Context, req *csi.NodePublis newVolume, err := ns.stageNewVolume(volumeID, stagingTargetPath, volContext, readOnly) if err != nil { + ns.removeVolumeMutex(volumeID) return nil, status.Errorf(codes.Internal, "failed to re-stage volume: %v", err) } |