Adding a ginkgo version of monitoring e2e test #5332
Conversation
|
@zmerlynn: I have enabled monitoring by default for e2e as well. Can you confirm that the changes in this PR will work for GKE based e2e clusters as well? |
|
@vishh: This isn't enough for GKE (we don't use But furthermore, I think you may just need to conditionalize the test on provider == gce anyways? |
|
@zmerlynn: Both of your comments are valid. I have updated the test to only look for the monitoring pods. The test will also run only on GCE. |
vishh
added
the
priority/critical-urgent
| influxdbPW = "root" | ||
| podlistQuery = "select distinct(pod) from stats" | ||
| nodelistQuery = "select distinct(hostname) from machine" | ||
| sleepBetweenAttempts = 30 * time.Second |
There was a problem hiding this comment.
Seems a bit long, can we make it 5s? Is there a downside to do it more often?
There was a problem hiding this comment.
I reduced the timeout to 5 minutes and sleep duration to 10s. If this test is run as soon as the cluster is setup, it can take some time for the pods to be active.
There was a problem hiding this comment.
Also once I implement the wait for pods to be running, this timeout can be reduced to a minute.
|
@vmarmol: Thanks for the review. Addressed your comments. |
|
LGTM |
Adding a ginkgo version of monitoring e2e test
This e2e is more detailed than its shell counterpart. This PR removes the shell based monitoring e2e test.
Fixes #5212 #4186