You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@logging.apache.org by Matt Sicker <bo...@gmail.com> on 2020/08/29 19:52:21 UTC

[Log4j] Test failures in master

It seems like all the Equinox OSGi tests are failing with an error
related to Felix. I'd imagine this is related to a recent dependency
update for one or the other. Any ideas on how to fix that? Or should
we just revert that dependency update for now?

-- 
Matt Sicker <bo...@gmail.com>

Re: [Log4j] Test failures in master

Posted by Matt Sicker <bo...@gmail.com>.
It mentions a class signing error of some sorts, and there's a comment
in log4j-osgi/pom.xml that says:

Place Felix before Equinox because Felix is signed. / also place it
before org.osgi.core so that its versions of the OSGi classes are used

Seems like some other updates might have a similar problem.

On Sat, 29 Aug 2020 at 15:00, Ralph Goers <ra...@dslextreme.com> wrote:
>
> That depends on what the error is. In general fixing errors is better than sweeping them under the rug. I am not sure when I will get a chance to look into it myself.
>
> Ralph
>
> > On Aug 29, 2020, at 12:52 PM, Matt Sicker <bo...@gmail.com> wrote:
> >
> > It seems like all the Equinox OSGi tests are failing with an error
> > related to Felix. I'd imagine this is related to a recent dependency
> > update for one or the other. Any ideas on how to fix that? Or should
> > we just revert that dependency update for now?
> >
> > --
> > Matt Sicker <bo...@gmail.com>
> >
>
>


-- 
Matt Sicker <bo...@gmail.com>

Re: [Log4j] Test failures in master

Posted by Ralph Goers <ra...@dslextreme.com>.
That depends on what the error is. In general fixing errors is better than sweeping them under the rug. I am not sure when I will get a chance to look into it myself.

Ralph

> On Aug 29, 2020, at 12:52 PM, Matt Sicker <bo...@gmail.com> wrote:
> 
> It seems like all the Equinox OSGi tests are failing with an error
> related to Felix. I'd imagine this is related to a recent dependency
> update for one or the other. Any ideas on how to fix that? Or should
> we just revert that dependency update for now?
> 
> -- 
> Matt Sicker <bo...@gmail.com>
>