actions/actions-runner-controller
1
0
Fork 0
mirror of https://github.com/actions/actions-runner-controller.git synced 2025-12-10 19:50:30 +00:00
Code Issues Packages Projects Releases Wiki Activity

Make logging more concise (#410)

Browse Source 
This makes logging more concise by changing logger names to something like `controllers.Runner` to `actions-runner-controller.runner` after the standard `controller-rutime.controller` and reducing redundant logs by removing unnecessary requeues. I have also tweaked log messages so that their style is more consistent, which will also help readability. Also, runnerreplicaset-controller lacked useful logs so I have enhanced it.
This commit is contained in:
Yusuke Kuoka
2021-03-20 07:34:25 +09:00
committed by GitHub
parent 07f822bb08
commit bc6e499e4f
4 changed files with 60 additions and 43 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
controllers/runner_controller.go
View File

@@ -167,9 +167,11 @@ func (r *RunnerReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
// Gracefully handle pod-already-exists errors due to informer cache delay.
// Without this we got a few errors like the below on new runner pod:
// 2021-03-16T00:23:10.116Z ERROR controller-runtime.controller Reconciler error {"controller": "runner-controller", "request": "default/example-runnerdeploy-b2g2g-j4mcp", "error": "pods \"example-runnerdeploy-b2g2g-j4mcp\" already exists"}
log.Info("Runner pod already exists. Probably this pod has been already created in previous reconcilation but the new pod is not yet cached.")
log.Info(
"Failed to create pod due to AlreadyExists error. Probably this pod has been already created in previous reconcilation but is still not in the informer cache. Will retry on pod created. If it doesn't repeat, there's no problem",
)
return ctrl.Result{RequeueAfter: 10 * time.Second}, nil
return ctrl.Result{}, nil
}
log.Error(err, "Failed to create pod resource")
@@ -187,7 +189,7 @@ func (r *RunnerReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
if deletionDidTimeout {
log.Info(
"Pod failed to delete itself in a timely manner. "+
fmt.Sprintf("Failed to delete pod within %s. ", deletionTimeout)+
"This is typically the case when a Kubernetes node became unreachable "+
"and the kube controller started evicting nodes. Forcefully deleting the pod to not get stuck.",
"podDeletionTimestamp", pod.DeletionTimestamp,
@@ -262,9 +264,12 @@ func (r *RunnerReconciler) Reconcile(req ctrl.Request) (ctrl.Result, error) {
if lastCheckTime := runner.Status.LastRegistrationCheckTime; lastCheckTime != nil {
nextCheckTime := lastCheckTime.Add(registrationCheckInterval)
now := time.Now()
if nextCheckTime.After(now) {
requeueAfter := nextCheckTime.Sub(now)
// Requeue scheduled by RequeueAfter can happen a bit earlier (like dozens of milliseconds)
// so to avoid excessive, in-effective retry, we heuristically ignore the remaining delay in case it is
// shorter than 1s
requeueAfter := nextCheckTime.Sub(now) - time.Second
if requeueAfter > 0 {
log.Info(
fmt.Sprintf("Skipped registration check because it's deferred until %s. Retrying in %s at latest", nextCheckTime, requeueAfter),
"lastRegistrationCheckTime", lastCheckTime,
@@ -795,8 +800,6 @@ func (r *RunnerReconciler) SetupWithManager(mgr ctrl.Manager) error {
r.Recorder = mgr.GetEventRecorderFor(name)
r.Recorder = mgr.GetEventRecorderFor(name)
return ctrl.NewControllerManagedBy(mgr).
For(&v1alpha1.Runner{}).
Owns(&corev1.Pod{}).