-
Notifications
You must be signed in to change notification settings - Fork 2.7k
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
Enable test which now works because of serialization changes #40749
Enable test which now works because of serialization changes #40749
Conversation
Relates to quarkusio#40601, quarkusio#40749, quarkusio#38991. No use waiting for Dependabot. Fixes problem where an NPE can occur in some tests.
This comment has been minimized.
This comment has been minimized.
The failing tests are in my new test, so it looks like I need to be a bit more successful disabling it on native. Edit: Now fixed. |
abb8e73
to
41474ac
Compare
41474ac
to
b24844c
Compare
Status for workflow
|
@@ -1,193 +0,0 @@ | |||
# Log settings | |||
# log level in lower case for testing | |||
quarkus.log.level=info |
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 is this file removed? It was useless?
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.
When I initially wrote these tests, I'd copied the files from another test. It had a whole bunch of settings that we didn't actually want in this test, such as setting the non-application-root to 1234. (Although, as mentioned above, even without the settings file, the non-application root is still 1234, which I haven't managed to explain.)
I left the file, so we had change history, but removed all of the settings since we didn't need any of them, as far as I know.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
…#40749" This reverts commit fc3988b. It has some additional changes which re-revert part of quarkusio#40601, to reintroduce removed guards to avoid cloning things like Quarkus runtime classes. Otherwise we get test failures.
#35124 added several tests which could not pass, because of our test classloading issues. It's taken me a while (cough) to get on top of the code changes to make them pass, but @dmlloyd has fixed one category of them (and also quarkiverse/quarkus-pact#73) with #40601.
Because the test has been disabled for a while, it rotted slightly and got broken by some dev ui changes. The
DevUIJsonRPCTest
was seeing a non-defaultquarkus.http.non-application-root-path
, so things failed. I can’t see where the problem value is coming from; perhaps some sticky config changes from a previous test? Cross-talk between tests is bad, but I didn’t want to get sucked into a config rabbit hole, so I just set my application to have the same config as the intruding config and moved on.I also confirmed that #27821 and #22611 are still problems (and patched up their tests to fail for the right reasons again, so they will hopefully pass when the epic test classloading rewrite ETCR (#34681) is done.