Skip to content

backport: storage/raft: Fix leadership deadlock (#8547) #8563

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
briankassouf merged 1 commit into from
Mar 13, 2020
Merged

backport: storage/raft: Fix leadership deadlock (#8547) #8563

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
19 changes: 14 additions & 5 deletions physical/raft/raft.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -999,12 +999,21 @@ type RaftLock struct {
func (l *RaftLock) monitorLeadership(stopCh <-chan struct{}, leaderNotifyCh <-chan bool) <-chan struct{} {
leaderLost := make(chan struct{})
go func() {
select {
case isLeader := <-leaderNotifyCh:
if !isLeader {
close(leaderLost)
for {
select {
case isLeader := <-leaderNotifyCh:
// leaderNotifyCh may deliver a true value initially if this
// server is already the leader prior to RaftLock.Lock call
// (the true message was already queued). The next message is
// always going to be false. The for loop should loop at most
// twice.
if !isLeader {
close(leaderLost)
return
}
case <-stopCh:
return
}
case <-stopCh:
}
}()
return leaderLost
Expand Down