feat: add keep-running-on-failure-implementation#8270
Conversation
Codecov Report
@@ Coverage Diff @@
## main #8270 +/- ##
==========================================
- Coverage 70.48% 66.01% -4.48%
==========================================
Files 515 605 +90
Lines 23150 29694 +6544
==========================================
+ Hits 16317 19602 +3285
- Misses 5776 8620 +2844
- Partials 1057 1472 +415
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
b6059dc to
8b3d916
Compare
225c76e to
18f9fd3
Compare
18f9fd3 to
bc98610
Compare
|
Just tried it out, seems to be working well for most cases! I noticed though that the error is not necessarily always the same as what would be output w/o the --keep-running-on-failure flag. For example if i have a remote kubecontext: And I attempt If I run this with Is there a way that we can preserve these actionable errors that we surface, it seems currently using this flag can sometimes mask more actionable error messages that we would normally surface |
|
Also we should create a new feature-request github issue around having skaffold surface the compiler error (docker error logs) if that can be grabbed from the docker build logs when using |
| var err error | ||
| bRes, err := r.Build(ctx, out, artifacts) | ||
| for ; err != nil && r.runCtx.Opts.KeepRunningOnFailure; bRes, err = r.Build(ctx, out, artifacts) { | ||
| log.Entry(ctx).Warnf("Failed to build artifacts, please fix the error and press any key to continue.") |
There was a problem hiding this comment.
you need to print the err as well
| if r.runCtx.IsTestPhaseActive() { | ||
| err = r.Test(ctx, out, bRes) | ||
| for ; err != nil && r.runCtx.Opts.KeepRunningOnFailure; err = r.Test(ctx, out, bRes) { | ||
| log.Entry(ctx).Warnf("Failed to run tests, please fix the error and press any key to continue.") |
There was a problem hiding this comment.
same here, the err message should be outputted
There was a problem hiding this comment.
ahhh.. you're right! I'll add err message.. Thanks!!
|
I have a case where log tailing stops.
|
@gsquared94 great catch! Thank you!
|
fixes: #4158
Description
updating any files, so file changes can be captured.runs, the deploy is not happening. I think the root problem might be that we set intent immediately after the skaffold server receives the intent, we could set the intent in skaffold session when the listener gets intent from the intent channel, that might require large change, so not doing it in this pr.Test Plan
skaffold dev --keep-running-on-failureWARN[0008] Failed to build artifacts, please fix the error and press any key to continue. subtask=-1 task=DevLoopfrom console