-
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
Changed the usage from slf4j-jboss-logging to slf4j-jboss-logmanager. #12692
Changed the usage from slf4j-jboss-logging to slf4j-jboss-logmanager. #12692
Conversation
@SlyngDK I've released 1.1.0.Final of the |
@dmlloyd I started to see the
|
Are you able to attach a debugger to this, to check that all three fields are non-null at this point (thus showing that all three attachments are in fact populated and there isn't some bug I'm not seeing)? If so I guess I'll add one or two more slots, and release again. |
|
That's odd. The keys are different, but the values are the same. The other odd part is two of them have a |
The values aren't identical - I bet we're seeing the logger be added from two different class loaders. Looking at |
I bet it's a test class loader issue. If I run the test by itself it passes. It's the third test that fails too and there are 3 attachments allowed. |
That is it. If I used:
I don't see the error message. |
In which pom.xml, should that be added? |
I found another possible solution.
to |
@SlyngDK I set it in the vert.x core tests since it was the first one that was failing. I really just did that to confirm my theory because as you said it slows tests down quite a bit :) |
The build was successful and ready to review. |
I squashed the commits. |
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.
Questions:
- should we do the same for the others?
commons-logging
for instance? - should we update the documentation we have here: https://quarkus.io/guides/logging#logging-adapters ?
- should we also add exclusions for the other
-jboss-logging
artifacts (we already recommend using the logmanager for Log4j and Log4j2)? - can we check that the new components used are Apache licensed? If not, can we release them under the Apache License.
I'm done with my questions :). I know some of these questions are broader than the current PR but let's get it right for all the adapters.
@vsevel could you test the failure you reported with this PR? Thanks! |
The docs in
I can add them if you want it.
|
I can confirm that |
My question is not if it's not an issue to use the logmanager artifact or not but more: is it the right thing to do? From what I can see, it looks like the right thing to do so I think we should be consistent. Except if I'm missing something? |
My apologies @gsmet. Yes I think using the |
@jamezp no need to apologize, I'm just trying to understand the right thing to do :). So IIUC, we should use the |
@gsmet Looking at the bom, hopefully that's the correct place, it looks like once the slf4j-jboss-logging dependency is replaced with the slf4j-jboss-logmanager version then the only left would be |
@dmlloyd could you shed some light on James' last comment in the context of Quarkus? If possible, I would like us to get it right and be as consistent as possible. So if it's a possibility we use the |
Sure, it can be done either way. If we want to test the other library though, let's do it now so I know if I need to make any other fixes. |
@gsmet What work need to be done on this PR, before it can be merged? Is there something I can do? |
Merged, thanks! |
Is related with #12615
To be able to support forwarding slf4j logging parameters to LogManager without formatting more than once.
The change from
slf4j-jboss-logging
toslf4j-jboss-logmanager
.A new release is still required of
slf4j-jboss-logmanager
, to include the parameter forwarding.