gsh-digital-services/seaweedFS
5
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

fix(master): fast resume state and default resumeState to true (#8925)

Browse Source 
* fix(master): fast resume state and default resumeState to true

When resumeState is enabled in single-master mode, the raft server had
existing log entries so the self-join path couldn't promote to leader.
The server waited the full election timeout (10-20s) before self-electing.

Fix by temporarily setting election timeout to 1ms before Start() when
in single-master + resumeState mode with existing log, then restoring
the original timeout after leader election. This makes resume near-instant.

Also change the default for resumeState from false to true across all
CLI commands (master, mini, server) so state is preserved by default.

* fix(master): prevent fastResume goroutine from hanging forever

Use defer to guarantee election timeout is always restored, and bound
the polling loop with a timeout so it cannot spin indefinitely if
leader election never succeeds.

* fix(master): use ticker instead of time.After in fastResume polling loop
This commit is contained in:
Chris Lu
2026-04-04 14:15:56 -07:00
committed by GitHub
parent 0da1794856
commit 4efe0acaf5
4 changed files with 34 additions and 3 deletions
Download Patch File Download Diff File Expand all files Collapse all files

30
weed/server/raft_server.go
View File

@@ -32,6 +32,7 @@ type RaftServerOption struct {
DataDir string
Topo *topology.Topology
RaftResumeState bool
SingleMaster bool
HeartbeatInterval time.Duration
ElectionTimeout time.Duration
RaftBootstrap bool
@@ -176,10 +177,38 @@ func NewRaftServer(option *RaftServerOption) (*RaftServer, error) {
if err := s.raftServer.LoadSnapshot(); err != nil {
return nil, err
}
// In single-master mode resuming state, the log is not empty so the
// normal self-join path won't promote to leader. The server will
// self-elect after the election timeout, so use a tiny timeout to
// make this near-instant, then restore the original after election.
fastResume := option.SingleMaster && option.RaftResumeState && !s.raftServer.IsLogEmpty()
if fastResume {
s.raftServer.SetElectionTimeout(time.Millisecond)
}
if err := s.raftServer.Start(); err != nil {
return nil, err
}
if fastResume {
go func() {
defer s.raftServer.SetElectionTimeout(option.ElectionTimeout)
ticker := time.NewTicker(100 * time.Millisecond)
defer ticker.Stop()
timeout := time.After(option.ElectionTimeout)
for s.raftServer.Leader() == "" {
select {
case <-timeout:
glog.Warningf("Fast resume timed out waiting for leader election, restoring election timeout to %v", option.ElectionTimeout)
return
case <-ticker.C:
}
}
glog.V(0).Infof("Resumed as leader with election timeout restored to %v", option.ElectionTimeout)
}()
}
for name, peer := range s.peers {
if err := s.raftServer.AddPeer(name, peer.ToGrpcAddress()); err != nil {
return nil, err
@@ -273,3 +302,4 @@ func (s *RaftServer) DoJoinCommand() {
}
}