Open yujuhong opened 8 years ago
@yujuhong do you have a list of items that you could federate? In general, the responsiveness of the kubelet is an issue.
/cc @kubernetes/rh-cluster-infra @rrati
@yujuhong is this still in plan for 1.4?
This needs to be triaged as a release-blocker or not for 1.5 @yujuhong
I had a PR for parallelizing inspections but I couldn't observe significant differences in performances in my measurements. I think we need to benchmark/profile kubelet to identify the bottleneck first.
I am removing the milestone and mark this as backlog. Will raise the priority if this becomes a serious issue.
Issues go stale after 90d of inactivity.
Mark the issue as fresh with /remove-lifecycle stale
.
Stale issues rot after an additional 30d of inactivity and eventually close.
Prevent issues from auto-closing with an /lifecycle frozen
comment.
If this issue is safe to close now please do so with /close
.
Send feedback to sig-testing, kubernetes/test-infra and/or @fejta
.
/lifecycle stale
/triage accepted
This issue has not been updated in over 1 year, and should be re-triaged.
You can:
/triage accepted
(org members only)/close
For more details on the triage process, see https://www.kubernetes.dev/docs/guide/issue-triage/
/remove-triage accepted
This issue is currently awaiting triage.
If a SIG or subproject determines this is a relevant issue, they will accept it by applying the triage/accepted
label and provide further guidance.
The triage/accepted
label can be added by org members by writing /triage accepted
in a comment.
Forked from https://github.com/kubernetes/kubernetes/issues/23591#issuecomment-203042820 Creating a new issue so I won't forget.
PLEG serially gets status for every pod, where at least one container has encountered a state transition. This becomes the bottleneck if a lot of containers changed in one relist period. We were aware of this problem and had come up with a few options before:
docker ps -a
call since pleg already have this information. The downside is that this is a very docker-specific optimization and doesn't make sense for other runtimes.I think we can do (2) with a small number of goroutines (e.g., 2) to speed up the time.