-
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
Optimize config generation #32209
Comments
You added a link to a Zulip discussion, please make sure the description of the issue is comprehensive and doesn't require accessing Zulip
|
I will experiment with this soon and report back. |
#32209 changes RESTEasy Reactive configuration to @ConfigMapping |
Oops :) |
I don't think this is done yet on "full scale". I'll re-open it for further evaluation. |
Thanks for catching this, yes GitHub auto-closed this when I merged the PR but that's wrong. |
It would be interesting to measure if the changes are showing any impact. I'm pretty sure they will, but nothing like confirming it :) |
One way of testing an impact would be re-running the test mentioned in the description with |
FWIW, I just tried to lower the memory for these tests again and things look worse than before as I can't even get to the point of starting Quarkus: the bootstrap itself requires more than 128 MB of heap space. Failure is in:
|
/cc @aloubyansky ^ |
Above stacktrace is testing:
|
When it comes to TestMojoIT, there is only one test that loads only 4 workspace modules. |
From what I can see I have failure in the Failsafe tests also if I disable Surefire. |
Yes. I reduced it to 64M first to see how it was doing, then raised it to 128M, which was our initial value, before we started having issues IIRC. |
I ran |
Ah, hold on, I'll re-test. |
Yes, still OOME. |
It doesn't look like it's the size of the workspace though, it loads only for modules. It must be something else. |
I haven't tried to dump the memory on OOME. Do you want me to do that? |
Sure, thanks. |
We did many optimizations in the last few releases with:
And many more... I think we can close this one. |
Description
After merging #30506
io.quarkus.maven.it.DevMojoIT.testThatTheApplicationIsReloadedMultiModule
started to fail with the OOME. Although these dev mode tests are run with-Xmx128m
, they do highlight an issue in the way our config classes are generated currently.Implementation ideas
As an alternative to the current impl, @radcortez suggested to switch to the ConfigMapping approach instead in https://quarkusio.zulipchat.com/#narrow/stream/187038-dev/topic/OOME.20in.20main/near/343476381
The text was updated successfully, but these errors were encountered: