Skip to content

fix: Ensure coordinator is closed and freed in agent #4164

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
merged 2 commits into from
Sep 23, 2022
Merged

fix: Ensure coordinator is closed and freed in agent #4164

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
d84d792
fix: Close coordinator on context cancellation
mafredri Sep 23, 2022
7e2ad50
fix: Refactor runCoordinator so that previous is closed/freed
mafredri Sep 23, 2022
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
27 changes: 18 additions & 9 deletions agent/agent.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -272,6 +272,15 @@ func (a *agent) runTailnet(ctx context.Context, derpMap *tailcfg.DERPMap) {

// runCoordinator listens for nodes and updates the self-node as it changes.
func (a *agent) runCoordinator(ctx context.Context) {
for {
reconnect := a.runCoordinatorWithRetry(ctx)
if !reconnect {
return
}
}
}

func (a *agent) runCoordinatorWithRetry(ctx context.Context) (reconnect bool) {
var coordinator net.Conn
var err error
// An exponential back-off occurs when the connection is failing to dial.
Expand All @@ -280,38 +289,38 @@ func (a *agent) runCoordinator(ctx context.Context) {
coordinator, err = a.coordinatorDialer(ctx)
if err != nil {
if errors.Is(err, context.Canceled) {
return
return false
}
if a.isClosed() {
return
return false
}
a.logger.Warn(context.Background(), "failed to dial", slog.Error(err))
continue
}
//nolint:revive // Defer is ok because we're exiting this loop.
defer coordinator.Close()
a.logger.Info(context.Background(), "connected to coordination server")
break
}
select {
case <-ctx.Done():
return
return false
default:
}
defer coordinator.Close()
sendNodes, errChan := tailnet.ServeCoordinator(coordinator, a.network.UpdateNodes)
a.network.SetNodeCallback(sendNodes)
select {
case <-ctx.Done():
return
return false
case err := <-errChan:
if a.isClosed() {
return
return false
}
if errors.Is(err, context.Canceled) {
return
return false
}
a.logger.Debug(ctx, "node broker accept exited; restarting connection", slog.Error(err))
a.runCoordinator(ctx)
return
return true
}
}

Expand Down