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.

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

Kubelet fails to provide stats for static pods #5688

Closed
vishh opened this issue Mar 20, 2015 · 10 comments · Fixed by #5730
Closed

Kubelet fails to provide stats for static pods #5688

vishh opened this issue Mar 20, 2015 · 10 comments · Fixed by #5730
Assignees
Labels
area/introspection priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. sig/node Categorizes an issue or PR as relevant to SIG Node.
Milestone

Comments

@vishh
Copy link
Contributor

vishh commented Mar 20, 2015

Since static pods are now exposed via the apiserver, users (heapster) naturally expect kubelet to provide stats for the static pods.
This is breaking heapster.

cc @yujuhong

@vishh vishh added priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. sig/node Categorizes an issue or PR as relevant to SIG Node. labels Mar 20, 2015
@yujuhong
Copy link
Contributor

I am not familiar with stats collection in kubelet. What type of status are you referring here?

@dchen1107 dchen1107 added this to the v1.0 milestone Mar 20, 2015
@yujuhong yujuhong self-assigned this Mar 20, 2015
@dchen1107
Copy link
Member

Not status, it is resource related stats here. This is one of initial reason I want to sync static pods back to master.

@yujuhong
Copy link
Contributor

I think we'll need to support pod translation for the following functions.

GetContainerInfo()
GetKubeletContainerLogs()

I will work on that.

@vmarmol
Copy link
Contributor

vmarmol commented Mar 20, 2015

On a side note, is Heapster breaking, breaking? Or just complaining loudly?

@vishh
Copy link
Contributor Author

vishh commented Mar 20, 2015

@vmarmol: Heapster fails stats collection all together if kubelet errors out. I have a filed a
heapster issue to fix that.

@dchen1107
Copy link
Member

@vishh how come the monitor e2e test couldn't detect such failure at the first place?

@vishh
Copy link
Contributor Author

vishh commented Mar 20, 2015 via email

@vishh
Copy link
Contributor Author

vishh commented Mar 20, 2015

One option is to run the monitoring test serially and then run other tests in parallel. @zmerlynn do you think it is possible to never run the monitoring e2e test in parallel with other e2e tests?

@zmerlynn
Copy link
Member

We currently don't run any e2e tests in parallel, but presumably we'll have to make such exceptions at some point. Why, though?

@vishh
Copy link
Contributor Author

vishh commented Mar 20, 2015

I intend the monitoring e2e test to ensure all the pods existing in the
cluster irrespective of , to be captured by the monitoring pipeline.

On Fri, Mar 20, 2015 at 9:28 AM, Zach Loafman notifications@github.com
wrote:

We currently don't run any e2e tests in parallel, but presumably we'll
have to make such exceptions at some point. Why, though?


Reply to this email directly or view it on GitHub
#5688 (comment)
.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area/introspection priority/important-soon Must be staffed and worked on either currently, or very soon, ideally in time for the next release. sig/node Categorizes an issue or PR as relevant to SIG Node.
Projects
None yet
Development

Successfully merging a pull request may close this issue.

5 participants