-
Notifications
You must be signed in to change notification settings - Fork 1.6k
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
fix(backend): Remove resource from workflow-controller-configmap. Closes #9087 #9089
Conversation
manifests/kustomize/third-party/argo/base/workflow-controller-configmap-patch.yaml
Show resolved
Hide resolved
/test kubeflow-pipelines-samples-v2 |
/retest |
2 similar comments
/retest |
/retest |
@gkcalat: The following test failed, say
Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. I understand the commands that are listed here. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
/lgtm
/approve
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: chensun The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
@chensun @gkcalat
I get a error message like this when running a XGBoost pipeline: |
Hi @jsitu777, Can you provide more info:
|
|
The workflow-controller-configmap after removing the quota:
|
Which version of Kubeflow are you using? |
Could you also provide the manifest of the failed pod
|
1.7.0 |
it will not describe. the pod is killed at the beginning. example: |
Another note is that if I do not specify quota in profile and also remove the quota part in workflow-controller configmap as in this PR. pipeline will run:
|
described the pod when finishes. Seems like the pipeline didn't specify a limits field, is it the reason that I should also not specify the limits.cpu and limits.memory in my profile?
|
@jsitu777 thank you for reporting this.
Adding I feel we may have better solutions:
Example apiVersion: v1
kind: LimitRange
metadata:
name: <user>-limit-range
spec:
limits:
- default:
cpu: 0.5
memory: 512Mi
defaultRequest:
cpu: 0.01
memory: 32Mi
type: Container Which can be applied a user namespace: kubectl apply -f <yaml> -n <user> /cc @chensun |
@gkcalat |
As we approach KFP v2 GA, we will update user documentation (including instructions on creating |
@gkcalat I tried to use the
same error message: I use the then dsl compile and upload the pipeline to run |
Take a look at the option 2 in #9089 (comment). The hard quota in Also, note that your containers must comply to whatever you set in |
Why are you comparing memory and cpu? I thought they are two different fields |
It was a typo. Corrected. |
So I apply the limit range and it's still not resolved:
I have also set my profile quota to be much higher:
|
never mind. I was setting the limit range in the wrong namespace.
|
Description of your changes:
Note: I made this as a separate PR for easier tracking
Issue: #9087
Changes:
workflow-controller-configmap
Context:
3.3.8
, the issue may no longer be relevantTests:
resources
inworkflow-controller-configmap
spec.resources{}
for other containerspython:3.7
)spec.resources{}
for other containersspec.resources{}
python:3.7
)spec.resources{}
for other containersChecklist: