Skip to content

Master backoff conditions fix. #82

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
wants to merge 2 commits into from
Closed

Master backoff conditions fix. #82

Changes from 1 commit
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
f17c7fd
Fix to update conditions in backoff.
dmatch01 Nov 16, 2020
0c2c1e4
Adjusted logging level in cleanup and fixed syntax error.
dmatch01 Nov 16, 2020
File filter

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
Fix to update conditions in backoff. 
Signed-off-by: Diana Arroyo <darroyo@us.ibm.com>
  • Loading branch information
@dmatch01
dmatch01 committed Nov 16, 2020
commit f17c7fd29f5848b543273fbc0a026f2e55f57e21
50 changes: 35 additions & 15 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 @@ -803,21 +803,41 @@ func (cc *XController) updateEtcd(qj *arbv1.AppWrapper, at string) error {
}

func (qjm *XController) backoff(q *arbv1.AppWrapper, reason string, message string) {
q.Status.QueueJobState = arbv1.AppWrapperCondBackoff
cond := GenerateAppWrapperCondition(arbv1.AppWrapperCondBackoff, v1.ConditionTrue, reason, message)
q.Status.Conditions = append(q.Status.Conditions, cond)
q.Status.FilterIgnore = true // update QueueJobState only, no work needed
qjm.updateEtcd(q, "[backoff]Rejoining")
qjm.qjqueue.AddUnschedulableIfNotPresent(q)
glog.V(3).Infof("[backoff] %s move to unschedulableQ before sleep for %d seconds. activeQ=%t Unsched=%t &qj=%p Version=%s Status=%+v", q.Name, qjm.serverOption.BackoffTime, qjm.qjqueue.IfExistActiveQ((q)), qjm.qjqueue.IfExistUnschedulableQ((q)), q, q.ResourceVersion, q.Status)
time.Sleep(time.Duration(qjm.serverOption.BackoffTime) * time.Second)
qjm.qjqueue.MoveToActiveQueueIfExists(q)
q.Status.QueueJobState = arbv1.AppWrapperCondQueueing
returnCond := GenerateAppWrapperCondition(arbv1.AppWrapperCondQueueing, v1.ConditionTrue, "BackoffTimerExpired.", "")
q.Status.Conditions = append(q.Status.Conditions, returnCond)
q.Status.FilterIgnore = true // update QueueJobState only, no work needed
qjm.updateEtcd(q, "[backoff] Queueing")
glog.V(3).Infof("[backoff] %s activeQ.Add after sleep for %d seconds. activeQ=%t Unsched=%t &qj=%p Version=%s Status=%+v", q.Name, qjm.serverOption.BackoffTime, qjm.qjqueue.IfExistActiveQ((q)), qjm.qjqueue.IfExistUnschedulableQ((q)), q, q.ResourceVersion, q.Status)
var workingAW *arbv1.AppWrapper
apiCacheAWJob, e := qjm.queueJobLister.AppWrappers(q.Namespace).Get(q.Name)
// Update condition
if (e == nil) {
workingAW = apiCacheAWJob
apiCacheAWJob.Status.QueueJobState = arbv1.AppWrapperCondBackoff
cond := GenerateAppWrapperCondition(arbv1.AppWrapperCondBackoff, v1.ConditionTrue, reason, message)
workingAW.Status.Conditions = append(workingAW.Status.Conditions, cond)
workingAW.Status.FilterIgnore = true // update QueueJobState only, no work needed
qjm.updateEtcd(workingAW, "backoff - Rejoining")
} else {
workingAW = q
glog.Errorf("[backoff] Failed to retrieve cached object for %s/%s. Continuing with possible stale object without updating conditions.", workingAW.Namespace,workingAW.Name)
}
qjm.qjqueue.AddUnschedulableIfNotPresent(workingAW)
glog.V(3).Infof("[backoff] %s move to unschedulableQ before sleep for %d seconds. activeQ=%t Unsched=%t &qj=%p Version=%s Status=%+v", workingAW.Name,
qjm.serverOption.BackoffTime, qjm.qjqueue.IfExistActiveQ((workingAW)), qjm.qjqueue.IfExistUnschedulableQ((workingAW)), workingAW, workingAW.ResourceVersion, workingAW.Status)
time.Sleep(time.Duration(qjm.serverOption.BackoffTime) * time.Second)
qjm.qjqueue.MoveToActiveQueueIfExists(workingAW)

// Update condition after backoff
apiCacheAWJob, e = qjm.queueJobLister.AppWrappers(q.Namespace).Get(q.Name)
if (e == nil) {
workingAW = apiCacheAWJob
workingAW.Status.QueueJobState = arbv1.AppWrapperCondQueueing
returnCond := GenerateAppWrapperCondition(arbv1.AppWrapperCondQueueing, v1.ConditionTrue, "BackoffTimerExpired.", "")
workingAW.Status.Conditions = append(workingAW.Status.Conditions, returnCond)
workingAW.Status.FilterIgnore = true // update QueueJobState only, no work needed
qjm.updateEtcd(workingAW, "backoff - Queueing")
}
glog.V(3).Infof("[backoff] %s activeQ.Add after sleep for %d seconds. activeQ=%t Unsched=%t &aw=%p Version=%s Status=%+v", workingAW.Name,
qjm.serverOption.BackoffTime, qjm.qjqueue.IfExistActiveQ((workingAW)), qjm.qjqueue.IfExistUnschedulableQ((workingAW)), workingAW, workingAW.ResourceVersion, workingAW.Status)
}

// Run start AppWrapper Controller
}

// Run start AppWrapper Controller
Expand Down