Skip to content

OCPBUGS-85677: avoid Progressing during node reboot pod recreation #3018

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
jluhrsen wants to merge 1 commit into from
Closed

OCPBUGS-85677: avoid Progressing during node reboot pod recreation #3018

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: 8 additions & 6 deletions pkg/controller/statusmanager/pod_status.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -94,18 +94,19 @@ func (status *StatusManager) SetFromPods() {
for _, ds := range daemonSets {
dsName := NewClusteredName(ds)
dsState, hadState := daemonsetStates[dsName]
dsRolloutActive := !status.installComplete || ds.Status.UpdatedNumberScheduled < ds.Status.CurrentNumberScheduled
dsRolloutPending := ds.Generation > ds.Status.ObservedGeneration
dsRolloutActive := !status.installComplete || dsRolloutPending || (hadState && ds.Status.UpdatedNumberScheduled < ds.Status.CurrentNumberScheduled)

dsProgressing := false

if isNonCritical(ds) && ds.Status.NumberReady == 0 && !status.installComplete {
progressing = append(progressing, fmt.Sprintf("DaemonSet %q is waiting for other operators to become ready", dsName.String()))
dsProgressing = true
} else if ds.Status.UpdatedNumberScheduled < ds.Status.CurrentNumberScheduled {
} else if ds.Status.UpdatedNumberScheduled < ds.Status.CurrentNumberScheduled && dsRolloutActive {
progressing = append(progressing, fmt.Sprintf("DaemonSet %q update is rolling out (%d out of %d updated)", dsName.String(), ds.Status.UpdatedNumberScheduled, ds.Status.CurrentNumberScheduled))
dsProgressing = true
} else if ds.Status.NumberUnavailable > 0 {
if dsRolloutActive {
if dsRolloutActive || hadState {
progressing = append(progressing, fmt.Sprintf("DaemonSet %q is not available (awaiting %d nodes)", dsName.String(), ds.Status.NumberUnavailable))
dsProgressing = true
}
Expand All @@ -123,7 +124,8 @@ func (status *StatusManager) SetFromPods() {

var dsHung *string

if dsProgressing && !isNonCritical(ds) {
trackDSRollout := (dsProgressing && !isNonCritical(ds)) || dsRolloutPending
if trackDSRollout {
reachedAvailableLevel = false

if !hadState || !reflect.DeepEqual(dsState.LastSeenStatus, ds.Status) {
Expand All @@ -132,13 +134,13 @@ func (status *StatusManager) SetFromPods() {
}

// Catch hung rollouts
if hadState && (time.Since(dsState.LastChangeTime)) > ProgressTimeout {
if dsProgressing && !isNonCritical(ds) && hadState && (time.Since(dsState.LastChangeTime)) > ProgressTimeout {
hung = append(hung, fmt.Sprintf("DaemonSet %q rollout is not making progress - last change %s", dsName.String(), dsState.LastChangeTime.Format(time.RFC3339)))
empty := ""
dsHung = &empty
}
}
if dsProgressing && !isNonCritical(ds) {
if trackDSRollout {
daemonsetStates[dsName] = dsState
} else {
delete(daemonsetStates, dsName)
Expand Down