Skip to content

[manager/dispatcher] Fix deadlock in dispatcher shutdown #2742

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

Closed
anshulpundir wants to merge 1 commit into from
Closed

[manager/dispatcher] Fix deadlock in dispatcher shutdown #2742

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
14 changes: 7 additions & 7 deletions manager/dispatcher/dispatcher.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -342,6 +342,13 @@ func (d *Dispatcher) Stop() error {
d.cancel()
d.mu.Unlock()

d.processUpdatesLock.Lock()
// In case there are any waiters. There is no chance of any starting
// after this point, because they check if the context is canceled
// before waiting.
d.processUpdatesCond.Broadcast()
d.processUpdatesLock.Unlock()

// The active nodes list can be cleaned out only when all
// existing RPCs have finished.
// RPCs that start after rpcRW.Unlock() should find the context
Expand All @@ -351,13 +358,6 @@ func (d *Dispatcher) Stop() error {
d.downNodes.Clean()
d.rpcRW.Unlock()

d.processUpdatesLock.Lock()
// In case there are any waiters. There is no chance of any starting
// after this point, because they check if the context is canceled
// before waiting.
d.processUpdatesCond.Broadcast()
d.processUpdatesLock.Unlock()

d.clusterUpdateQueue.Close()

// TODO(anshul): This use of Wait() could be unsafe.
Expand Down