Skip to content
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

[YUNIKORN-2724] Improve the signature of methods notifyTaskComplete() and ensureAppAndTaskCreated() #873

Closed
wants to merge 15 commits into from
Closed

[YUNIKORN-2724] Improve the signature of methods notifyTaskComplete() and ensureAppAndTaskCreated() #873

Show file tree
Hide file tree
Changes from 12 commits
Commits
Show all changes
15 commits
Select commit Hold shift + click to select a range
89d6ecf
[YUNIKORN-2724] Improve the signature of methods notifyTaskComplete()…
ryankert01 Jul 8, 2024
e2104a5
Merge branch 'master' into YUNIKORN-2724
pbacsko Aug 1, 2024
0669b1b
Merge branch 'master' into YUNIKORN-2724
pbacsko Aug 1, 2024
d7ef717
Merge branch 'master' into YUNIKORN-2724
pbacsko Aug 1, 2024
c46c7a8
code review 1
ryankert01 Aug 8, 2024
9903652
code review 2
ryankert01 Aug 8, 2024
d74d0dc
cleanup ensureAppAndTaskCreated
ryankert01 Aug 8, 2024
7b48895
Merge remote-tracking branch 'origin/master' into YUNIKORN-2724
ryankert01 Aug 8, 2024
5721af1
very complex merge
ryankert01 Aug 8, 2024
0468d36
fix comficts
ryankert01 Aug 8, 2024
4be3aba
code review
ryankert01 Aug 9, 2024
6849dcc
simplify deleteYunikornPod
ryankert01 Aug 9, 2024
47ba6d5
code review
ryankert01 Aug 11, 2024
41e88ad
Merge branch 'apache:master' into YUNIKORN-2724
ryankert01 Aug 11, 2024
788d6d9
code review
ryankert01 Aug 11, 2024
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
58 changes: 28 additions & 30 deletions pkg/cache/context.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -295,41 +295,41 @@
}

func (ctx *Context) updateYuniKornPod(appID string, pod *v1.Pod) {
var app *Application
taskID := string(pod.UID)
if app = ctx.getApplication(appID); app != nil {
app := ctx.getApplication(appID)
if app != nil {
if task := app.GetTask(taskID); task != nil {
task.setTaskPod(pod)
}
}

// treat terminated pods like a remove
if utils.IsPodTerminated(pod) {
ctx.notifyTaskComplete(appID, taskID)
if app != nil {
ctx.notifyTaskComplete(app, taskID)
}
log.Log(log.ShimContext).Debug("Request to update terminated pod, removing from cache", zap.String("podName", pod.Name))
ctx.schedulerCache.RemovePod(pod)
return
}

if ctx.schedulerCache.UpdatePod(pod) {
// pod was accepted; ensure the application and task objects have been created
ctx.ensureAppAndTaskCreated(pod)
ctx.ensureAppAndTaskCreated(pod, app)
}
}

func (ctx *Context) ensureAppAndTaskCreated(pod *v1.Pod) {
// get app metadata
appMeta, ok := getAppMetadata(pod)
if !ok {
log.Log(log.ShimContext).Warn("BUG: Unable to retrieve application metadata from YuniKorn-managed Pod",
zap.String("namespace", pod.Namespace),
zap.String("name", pod.Name))
return
}

func (ctx *Context) ensureAppAndTaskCreated(pod *v1.Pod, app *Application) {
// add app if it doesn't already exist
app := ctx.GetApplication(appMeta.ApplicationID)
if app == nil {
// get app metadata
appMeta, ok := getAppMetadata(pod)
if !ok {
log.Log(log.ShimContext).Warn("BUG: Unable to retrieve application metadata from YuniKorn-managed Pod",
zap.String("namespace", pod.Namespace),
zap.String("name", pod.Name))
return

Check warning on line 331 in pkg/cache/context.go

View check run for this annotation

Codecov / codecov/patch

pkg/cache/context.go#L328-L331 

Added lines #L328 - L331 were not covered by tests
}
app = ctx.AddApplication(&AddApplicationRequest{
Metadata: appMeta,
})
Expand Down Expand Up @@ -435,9 +435,7 @@

func (ctx *Context) deleteYuniKornPod(pod *v1.Pod) {
if taskMeta, ok := getTaskMetadata(pod); ok {
if app := ctx.GetApplication(taskMeta.ApplicationID); app != nil {
ctx.notifyTaskComplete(taskMeta.ApplicationID, taskMeta.TaskID)
}
ctx.notifyTaskComplete(ctx.GetApplication(taskMeta.ApplicationID), taskMeta.TaskID)
}

log.Log(log.ShimContext).Debug("removing pod from cache", zap.String("podName", pod.Name))
Expand Down Expand Up @@ -873,19 +871,19 @@
return ctx.schedulerCache.StartPodAllocation(podKey, nodeID)
}

func (ctx *Context) notifyTaskComplete(appID, taskID string) {
log.Log(log.ShimContext).Debug("NotifyTaskComplete",
zap.String("appID", appID),
zap.String("taskID", taskID))
if app := ctx.GetApplication(appID); app != nil {
log.Log(log.ShimContext).Debug("release allocation",
zap.String("appID", appID),
func (ctx *Context) notifyTaskComplete(app *Application, taskID string) {
if app == nil {
log.Log(log.ShimContext).Debug("In notifyTaskComplete but app is nil",
zap.String("taskID", taskID))
ev := NewSimpleTaskEvent(appID, taskID, CompleteTask)
dispatcher.Dispatch(ev)
if app.GetApplicationState() == ApplicationStates().Resuming {
dispatcher.Dispatch(NewSimpleApplicationEvent(appID, AppTaskCompleted))
}
return
}
log.Log(log.ShimContext).Debug("notifyTaskComplete and release allocation",
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Maybe we can keep origin message? release allocation

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I change the message to "notifyTaskComplete and release allocation" because I combine 2 logs to make it only log once. Perhaps I can change it to " release allocation in notifyTaskComplete"?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Got it. It is fine to keep current message :)

zap.String("appID", app.applicationID),
zap.String("taskID", taskID))
ev := NewSimpleTaskEvent(app.applicationID, taskID, CompleteTask)
dispatcher.Dispatch(ev)
if app.GetApplicationState() == ApplicationStates().Resuming {
dispatcher.Dispatch(NewSimpleApplicationEvent(app.applicationID, AppTaskCompleted))

Check warning on line 886 in pkg/cache/context.go

View check run for this annotation

Codecov / codecov/patch

pkg/cache/context.go#L886 

Added line #L886 was not covered by tests
}
}

Expand Down
4 changes: 2 additions & 2 deletions pkg/cache/context_test.go
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1034,7 +1034,7 @@ func TestTaskReleaseAfterRecovery(t *testing.T) {
assert.Equal(t, len(app.GetBoundTasks()), 2)

// release one of the tasks
context.notifyTaskComplete(appID, pod2UID)
context.notifyTaskComplete(app, pod2UID)

// wait for release
err = utils.WaitForCondition(func() bool {
Expand Down Expand Up @@ -2075,7 +2075,7 @@ func TestTaskRemoveOnCompletion(t *testing.T) {
assert.NilError(t, err)

// mark completion
context.notifyTaskComplete(appID, taskUID1)
context.notifyTaskComplete(app, taskUID1)
err = utils.WaitForCondition(func() bool {
return task.GetTaskState() == TaskStates().Completed
}, 100*time.Millisecond, time.Second)
Expand Down