Skip to content

Quota management fix recover pending state #168

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
Aug 26, 2021
Merged

Quota management fix recover pending state #168

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
46b2c91
Fix to recovery of AppWrappers in Queue with pending state.
dmatch01 Aug 26, 2021
d391780
Increment minor ver. number.
dmatch01 Aug 26, 2021
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
2 changes: 1 addition & 1 deletion CONTROLLER_VERSION
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1 +1 @@
1.29.35
1.29.36
8 changes: 4 additions & 4 deletions pkg/controller/clusterstate/cache/cache.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -305,21 +305,21 @@ func (sc *ClusterStateCache) updateState() error {
}

klog.V(8).Infof("Total capacity %+v, used %+v, free space %+v", total, used, idle)
if klog.V(10).Enabled() {
if klog.V(12).Enabled() {
// CPU histogram
metricCPU := &dto.Metric{}
(*newIdleHistogram.MilliCPU).Write(metricCPU)
klog.V(10).Infof("[updateState] CPU histogram:\n%s", proto.MarshalTextString(metricCPU))
klog.V(12).Infof("[updateState] CPU histogram:\n%s", proto.MarshalTextString(metricCPU))

// Memory histogram
metricMem := &dto.Metric{}
(*newIdleHistogram.Memory).Write(metricMem)
klog.V(10).Infof("[updateState] Memory histogram:\n%s", proto.MarshalTextString(metricMem))
klog.V(12).Infof("[updateState] Memory histogram:\n%s", proto.MarshalTextString(metricMem))

// GPU histogram
metricGPU := &dto.Metric{}
(*newIdleHistogram.GPU).Write(metricGPU)
klog.V(10).Infof("[updateState] GPU histogram:\n%s", proto.MarshalTextString(metricGPU))
klog.V(12).Infof("[updateState] GPU histogram:\n%s", proto.MarshalTextString(metricGPU))
}

err := sc.saveState(idle, total, newIdleHistogram)
Expand Down
10 changes: 10 additions & 0 deletions pkg/controller/queuejob/queuejob_controller_ex.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1599,6 +1599,16 @@ func (cc *XController) manageQueueJob(qj *arbv1.AppWrapper, podPhaseChanges bool
return nil
} // End of first execution of qj to add to qjqueue for ScheduleNext

//Handle recovery condition
if !qj.Status.CanRun && qj.Status.State == arbv1.AppWrapperStateEnqueued &&
!cc.qjqueue.IfExistUnschedulableQ(qj) && !cc.qjqueue.IfExistActiveQ(qj){
cc.qjqueue.AddIfNotPresent(qj)
klog.V(3).Infof("[manageQueueJob] Recovered AppWrapper %s%s - added to active queue, Status=%+v",
qj.Namespace, qj.Name, qj.Status)

return nil
}

// add qj to Etcd for dispatch
if qj.Status.CanRun && qj.Status.State != arbv1.AppWrapperStateActive {
qj.Status.State = arbv1.AppWrapperStateActive
Expand Down