| Re: [simrel-dev] Fwd: [eclipse-platform/eclipse.platform.releng.aggregator] Release 4.37 (Issue #3245) |
If you're using the 4.38 builds then it problem is unrelated to respins.
Please have ongoing discussion on the issue rather than on the mailing list.
I hadn't switched the tests to 4.38 which indeed has a problem:
Which also means the Maven release is fine because the those were then signed with the certificate that expired 07-21 before it expired. That's that's a relieve and this problem is easier to fix.
I'll try to get to that later today.
Hi Ed
I think you have diagnosed a new hazard for Maven users. I have raised: eclipse-platform/eclipse.platform.releng.aggregator#3300
However AFAIAA when my Jenkins/Tycho build uses
<location includeAllPlatforms="false" includeConfigurePhase="false" includeMode="planner" includeSource="true" type="InstallableUnit">
<unit id="org.eclipse.sdk.feature.group" version="0.0.0"/>
<repository location="https://download.eclipse.org/eclipse/updates/I-builds"/>
</location>
the build is making direct use of the latest platform build. The problem my tests observe is therefore nothing to do with what is on Maven Central.
Following my usual Hudson/Buckminster practice: "Do not waste time debugging a build failure until it repeats". I have repeated the OCL master build with the same signer faiilure so instability in the I-builds does not seem like the explanation. See <https://ci.eclipse.org/ocl/job/ocl-master/1025/display/redirect>
Regards
Edward Willink
On 07/09/2025 08:52, Ed Merks via simrel-dev wrote:
According to the tests
these three all have the the same signer:
But, one thing I didn't consider is whether all three of them will actually be published as new versions to Maven Central? I guessing not, and indeed yes, this version is quite old:https://repo.maven.apache.org/maven2/org/eclipse/platform/org.eclipse.equinox.registry/
That makes me very sad. So, in the future, when the tests find a split-package-signing problem, we need to make sure that not just the qualifier changes but that all participants need at least a minor increment for that release cycle to ensure they are published for the next Maven publishing cycle.
In any case, this has nothing to do with a respin but rather the fact that the platform published to Maven Central yesterday so today you see the result of that. I suggest you open an issue here:
On 9/7/2025 9:02 AM, Edward Willink via simrel-dev wrote:
Hi
It looks as if the progress towards a re-spin is introducing a traditional problem.
See <https://ci.eclipse.org/ocl/job/ocl-master/1024/display/redirect>[ERROR] Run 1: HTMLExportOCLValidationResultTests.setUp:67->AbstractExportOCLValidationResultTests.setUp:85->AbstractValidityTestCase.setUp:288 » Security class "org.eclipse.core.runtime.RegistryFactory"'s signer information does not match signer information of other classes in the same packageRegards
Edward Willink
On 05/09/2025 09:47, Ed Merks via simrel-dev wrote:
FYI, it looks like a respin is coming from the Platform.
-------- Forwarded Message --------
Subject: Re: [eclipse-platform/eclipse.platform.releng.aggregator] Release 4.37 (Issue #3245) Date: Fri, 05 Sep 2025 01:25:47 -0700 From: Andrey Loskutov <notifications@xxxxxxxxxx> Reply-To: eclipse-platform/eclipse.platform.releng.aggregator <reply+AABS6TFDE3OYZIDMWGPGTWOGYZ7YXEVBNHHMPR7TY4@xxxxxxxxxxxxxxxx> To: eclipse-platform/eclipse.platform.releng.aggregator <eclipse.platform.releng.aggregator@xxxxxxxxxxxxxxxxxx> CC: Subscribed <subscribed@xxxxxxxxxxxxxxxxxx>
iloveeclipse left a comment (eclipse-platform/eclipse.platform.releng.aggregator#3245)
Note: we found a severe performance problem with JDT tooling due the patch integrated late in RC1.
The fix is merged on 4.38 master branch already and we are evaluating the possibility for 4.37 RC2a build.See more details at eclipse-jdt/eclipse.jdt.ui#2456
—
Reply to this email directly, view it on GitHub, or unsubscribe.
You are receiving this because you are subscribed to this thread.
_______________________________________________ simrel-dev mailing list simrel-dev@xxxxxxxxxxx To unsubscribe from this list, visit https://accounts.eclipse.org
_______________________________________________ simrel-dev mailing list simrel-dev@xxxxxxxxxxx To unsubscribe from this list, visit https://accounts.eclipse.org
_______________________________________________ simrel-dev mailing list simrel-dev@xxxxxxxxxxx To unsubscribe from this list, visit https://accounts.eclipse.org
