Closed Rajalakshmi-Girish closed 6 months ago
[APPROVALNOTIFIER] This PR is APPROVED
This pull-request has been approved by: mkumatag, Rajalakshmi-Girish
The full list of commands accepted by this bot can be found here.
The pull request process is described here
/lgtm
@Rajalakshmi-Girish: Updated the job-config
configmap in namespace prow
at cluster default
using the following files:
test-e2e-capi-ibmcloud-periodics.yaml
using file config/jobs/periodic/cluster-api-provider-ibmcloud/test-e2e-capi-ibmcloud-periodics.yaml
test-etcd-periodics.yaml
using file config/jobs/periodic/etcd/test-etcd-periodics.yaml
etcd-postsubmit.yaml
using file config/jobs/ppc64le-cloud/builds/etcd-postsubmit.yaml
@Rajalakshmi-Girish lets think about how to avoid such issues in the feature, can we add some linting to catch such issues? not sure if there is a way..
I'm not sure about this, will it make any difference when we just have a key without requests.. cc @Karthik-K-N
anyways this is a invalid syntax which needs a fix, hence merging.
All these days, etcd job has been sending only resources: {}
in job yaml due to this invalid syntax.
When I tested etcd flakes locally as well, most of the time the host had resource limitations.
Running the long pending TestWatchDelay
flake on a pod with required resources, displayed 0 failure for 30m of stress run.
@mkumatag @srivastav-abhishek
The indentation was wrong for resources metadata in jobs
test-e2e-capi-ibmcloud-periodics.yaml
,test-etcd-periodics.yaml
andetcd-postsubmit.yaml
, and hence no proper resources were allocated to the jobs. I see theTestWatchDelay
related flakiness improve to a major extent after proper resource allotment.