actions/actions-runner-controller
1
0
Fork 0
mirror of https://github.com/actions/actions-runner-controller.git synced 2025-12-11 03:57:01 +00:00
Code Issues Packages Projects Releases Wiki Activity

Handle offline runners gracefully (#341)

Browse Source 
* if a runner pod starts up with an invalid token, it will go in an 
infinite retry loop, appearing as RUNNING from the outside
* normally, this error situation is detected because no corresponding 
runner objects exists in GitHub and the pod will get removed after 
registration timeout
* if the GitHub runner object already existed before - e.g. because a 
finalizer was not properly run as part of a partial Kubernetes crash, 
the runner will always stay in a running mode, even updating the 
registration token will not kill the problematic pod
* introducing RunnerOffline exception that can be handled in runner 
controller and replicaset controller
* as runners are offline when a pod is completed and marked for restart, 
only do additional restart checks if no restart was already decided, 
making code a bit cleaner and saving GitHub API calls after each job 
completion
This commit is contained in:
Johannes Nicolai
2021-02-22 02:08:04 +01:00
committed by GitHub
parent dd0b9f3e95
commit 2d7fbbfb68
3 changed files with 94 additions and 48 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
controllers/runnerreplicaset_controller.go
View File

@@ -109,11 +109,15 @@ func (r *RunnerReplicaSetReconciler) Reconcile(req ctrl.Request) (ctrl.Result, e
busy, err := r.GitHubClient.IsRunnerBusy(ctx, runner.Spec.Enterprise, runner.Spec.Organization, runner.Spec.Repository, runner.Name)
if err != nil {
notRegistered := false
offline := false
var e *github.RunnerNotFound
if errors.As(err, &e) {
log.V(1).Info("Failed to check if runner is busy. Either this runner has never been successfully registered to GitHub or has not managed yet to, and therefore we prioritize it for deletion", "runnerName", runner.Name)
var notFoundException *github.RunnerNotFound
var offlineException *github.RunnerOffline
if errors.As(err, &notFoundException) {
log.V(1).Info("Failed to check if runner is busy. Either this runner has never been successfully registered to GitHub or it still needs more time.", "runnerName", runner.Name)
notRegistered = true
} else if errors.As(err, &offlineException) {
offline = true
} else {
var e *gogithub.RateLimitError
if errors.As(err, &e) {
@@ -140,7 +144,7 @@ func (r *RunnerReplicaSetReconciler) Reconcile(req ctrl.Request) (ctrl.Result, e
if notRegistered && registrationDidTimeout {
log.Info(
"Runner failed to register itself to GitHub in timely manner. "+
"Recreating the pod to see if it resolves the issue. "+
"Marking the runner for scale down. "+
"CAUTION: If you see this a lot, you should investigate the root cause. "+
"See https://github.com/summerwind/actions-runner-controller/issues/288",
"runnerCreationTimestamp", runner.CreationTimestamp,
@@ -150,6 +154,12 @@ func (r *RunnerReplicaSetReconciler) Reconcile(req ctrl.Request) (ctrl.Result, e
notBusy = append(notBusy, runner)
}
// offline runners should always be a great target for scale down
if offline {
notBusy = append(notBusy, runner)
}
} else if !busy {
notBusy = append(notBusy, runner)
}