-
Notifications
You must be signed in to change notification settings - Fork 25
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
[Ready for review] Fixed e2e tests #964
Conversation
7eaf23c
to
a9c2551
Compare
Should we fix those tests before merging this? Otherwise, regular e2e may always fail. |
We should also increase the timeout of istio, maybe to 12 min? As sometimes the istio job failed due to timeout |
a9c2551
to
1753338
Compare
d84fd5b
to
1306ca6
Compare
Yes, I fixed the tests. The vdb-gen test is still somewhat flaky, but it’s much more stable now. |
istio should be handled by other PR. This one is for leg4 and udx. Also, before increasing the timeout, we should confirm that Istio is failing due to slowness. For instance, issues with vdb-gen and udx are caused by other factors, so increasing the timeout would only extend our wait time without resolving the root problems. |
- name: Pick the name of the image | ||
uses: spilchen/switch-case-action@v2 | ||
id: legacy_vertica_image | ||
with: | ||
default: ghcr.io/${{ env.OWNER_LC }}/vertica-k8s:${{ github.sha }}-legacy | ||
default: docker.io/opentext/vertica-k8s-private:latest-legacy-master |
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.
What's the difference between the sha-legacy and the latest-legacy-master image?
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.
sha-legacy image will be built by e2e tests, and latest-legacy-master is built by devops and already exists before e2e tests run.
@@ -128,6 +128,9 @@ function scrutinizeForAdmintools() { | |||
OP_FILE="$ns.$v.scrutinize.tar" | |||
logInfo "Running scrutinize" | |||
set -o xtrace | |||
# scrutinize needs to be able to ssh to the pod | |||
POD_IP=$(kubectl get pod -n $ns $pod -o jsonpath='{.status.podIP}') | |||
kubectl exec -t -n $ns $pod -- bash -c "ssh-keyscan $POD_IP >> ~/.ssh/known_hosts" |
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.
Why do we need this now?
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.
Scrutinize still has a bug. See my comment under this Jira: https://jira.verticacorp.com/jira/browse/VER-96447. This is a workaround before client team fixed the scrutinize.
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.
Looks good!
This PR addresses issues in the daily run for leg-4 and udx.
Leg-4: I improved stability by refactoring and isolating vdb-gen, reducing the likelihood of failures. However, the root cause of the reconcile requeue not functioning in GitHub CI remains unclear. After we use our own runner, we can revisit this issue again. Additionally, I resolved K8s scrutinizer issues for other failed revive tests.
UDX: I updated the DevOps Jenkins pipeline and fixed a script issue in the e2e test. Now udx is stable.