refine gcp private and custom-dns presubmit jobs #67649
Conversation
|
/pj-rehearse periodic-ci-openshift-release-master-nightly-4.20-e2e-gcp-custom-dns |
|
@jianlinliu: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
/close |
|
@jianlinliu: Closed this PR. DetailsIn response to this:
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-sigs/prow repository. |
|
In favor of #67624 |
|
/reopen |
|
@jianlinliu: Reopened this PR. DetailsIn response to this:
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-sigs/prow repository. |
jianlinliu
changed the title
jianlinliu
force-pushed
the
gcp-custom-dns-rename
branch
5 times, most recently
from
d4f1f01 to
e831e1a
Compare
|
/pj-rehearse pull-ci-openshift-installer-release-4.19-gcp-private pull-ci-openshift-installer-release-4.20-gcp-custom-dns |
|
@jianlinliu: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
@jianlinliu: The following test failed, say
Full PR test history. Your PR dashboard. DetailsInstructions 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-sigs/prow repository. I understand the commands that are listed here. |
|
/pj-rehearse pull-ci-openshift-installer-release-4.20-gcp-custom-dns |
|
@jianlinliu: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
/pj-rhearse abort |
refine gcp private ci jobs
jianlinliu
force-pushed
the
gcp-custom-dns-rename
branch
from
e831e1a to
24a3384
Compare
|
[REHEARSALNOTIFIER]
Interacting with pj-rehearseComment: Once you are satisfied with the results of the rehearsals, comment: |
|
/pj-rehearse pull-ci-openshift-installer-release-4.20-gcp-custom-dns |
|
@jianlinliu: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
|
/pj-rehearse ack |
|
@jianlinliu: now processing your pj-rehearse request. Please allow up to 10 minutes for jobs to trigger or cancel. |
openshift-ci-robot
added
the
rehearsals-ack
|
cc @gpei and @yunjiang29 to review |
ci-operator/config/openshift-priv/installer/openshift-priv-installer-release-4.19.yaml
Show resolved
Hide resolved
|
/lgtm |
|
/lgtm |
| FEATURE_SET: CustomNoUpgrade | ||
| USER_PROVISIONED_DNS: "yes" | ||
| test: | ||
| - chain: cucushift-installer-check-cluster-health |
There was a problem hiding this comment.
So this will stop e2e tests from running on the custom DNS job?
Are we doing this because the e2e tests are permafailing? I am ok with that, but not sure if there are other options?
@sadasu & I were discussing this today and there were roughly ~10 tests that failed on both of the runs for the custom DNS job. I believe @gpei mentioned this before, and that it would not make sense to add ignores for each individual test.
There was a problem hiding this comment.
because the e2e tests are permafailing
This is one of the reasons. Some others:
- We set a public domain which resides in QE project, so it depends on QE's cluster profile. While the QE cloud project have a small number of budget. So skip the e2e testing to save the running time for saving the cloud cost.
- The job is for installer presubmits jobs, most of time, we are focus on installation results, so I drop the e2e to avoid the noise from unexpected e2e failure.
There was a problem hiding this comment.
The most ideal solution in practice would be to fix the tests in the origin repo, either by making them pass or skipping them, so that we can see the most accurate and reasonable e2e test results.
But I believe that having each team update their test code to adapt all test cases to every new cluster type we introduced would be a big workload for them, and potentially with a very lengthy timeline.
During our initial testing of custom DNS, I informed the Ingress team that certain tests were consistently failing. They submitted PR openshift/origin#29985 to fix these issues, but the PR is still under review to this day.
There was a problem hiding this comment.
@jianlinliu and @gpei thanks for that explanation. I agree that installer pre-submit jobs should can be limited to verifying a successful install.
We will be able to catch other errors during the periodic e2e custom-dns jobs. I am concerned about scenarios like openshift/origin#29985 and how that will affect the health of our periodic job within CR.
|
/approve |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: gpei, jianlinliu, patrickdillon, yunjiang29 The full list of commands accepted by this bot can be found here. The pull request process is described here DetailsNeeds approval from an approver in each of these files:
Approvers can indicate their approval by writing |
openshift-ci
bot
added
the
approved
|
LGTM thanks for working on this. |
openshift-merge-bot
bot
merged commit a923eaa
into
openshift:master
…67649) refine gcp private ci jobs
…67649) refine gcp private ci jobs
…67649) refine gcp private ci jobs
6 participants
run_if_changed: (gcp|google)on gcp custom-dns jobs, to help us catch regression early, e.g: CORS-4027: Updating feature gate name to GCPClusterHostedDNSInstall #67624 (comment)testto drop e2e testing from installer pre-submit gcp custom-dns jobs to save cloud cost.gcp-privatejob together to drop the useless steps.