-
Notifications
You must be signed in to change notification settings - Fork 263
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
e2e test failed #500
Comments
Adding notes here looking at the two referenced PR logs
That's failure to create the service. For #484 logs: Also all these tests are failing only for nightly release of serving. |
For the #495 logs, I'm not sure why the For the #484 logs, I'm pretty sure that this is related to knative/serving#5319 and maybe the way how BYO revision names are handled in Not sure if we are checking something wrong (and is fixed now in latest), or whether the upstream change is broken. |
/assign |
I think it's a timing issue. Although the service update is synchronous and kn returned when the config change is reconciled we see difference in revision-name for See below:
Here ^^ Now, get the status again after few seconds:
Here ^^ We can add a couple of seconds delay when we verify the expected vs actual targets to work around this in kn e2e. However, the service update returned only after the service conditions returned true and thereafter the revision name's value in target /cc @taragu @mattmoor We started seeing this in the nightly build since ~3 days, the related PR got into the serving master is knative/serving#5319. FYI this worked fine against serving v0.10.0. |
- for traffic splitting e2e tests - Check the comment on knative#500 (comment)
- for traffic splitting e2e tests - Check the comment on knative#500 (comment)
* Uses BYO revision names for traffic splitting e2e tests - Avoid getting the revision name after service create/update using LCR or LRR from service status * Pauses 3 seconds before verifying actual vs expected traffic targets - for traffic splitting e2e tests - Check the comment on #500 (comment)
Reopening.
|
Logged at knative/serving#6060 |
Related to knative#500 - The fix for serving#6060 merged (serving#6069), removing the workaround at client side e2e tests for additional LCR=LRR check after service update return. - Now we expect service status to have LCR=LRR if the service's conditions turned True after last update.
Related to #500 - The fix for serving#6060 merged (serving#6069), removing the workaround at client side e2e tests for additional LCR=LRR check after service update return. - Now we expect service status to have LCR=LRR if the service's conditions turned True after last update.
Close this issue cause it has been resolved a long time ago. |
…tive#500) * [SRVCLI-250] Add test-images and test-e2e-local cmds to Makefile * fix: Fix exec permission * fix: Fix tags as comma separated list * fix: Fix ko build for symlink dir
Affected PRs: #495 #484, which needs retest after fixing this.
Logs from e2e test:
The text was updated successfully, but these errors were encountered: