[jdk8u-dev] RFR: 8336451: [11u] GHA macos-13 and macos-15 builders are unable to resolve local hostname

[Severin Gehwolf]

On Fri, 23 Jan 2026 20:33:09 GMT, Severin Gehwolf <sgehwolf at openjdk.org> wrote:

> Another GHA work-around. Sometimes GHA tests on MacOSX fails with something like:
> 
> 
> VM initialization failed for: /Users/runner/jdk-macos-x64/jdk-1.8.0-internal+0_osx-x64_bin/j2sdk-image/jre/bin/java -classpath /Users/runner/work/jdk8u-dev/jdk8u-dev/test-results/testoutput/jdk_tier1/JTwork/classes/com/sun/jdi -Xdebug -Xrunjdwp:transport=dt_socket,address=sjc20-cw752-2deed029-f22c-4b58-be22-6c2b8db93551-8E6FDEE10350.local:49673,suspend=y AccessSpecifierTarg
> 
> 
> For test:
> 
> com/sun/jdi/RedefineCrossEvent.java
> 
> 
> See for example here:
> https://github.com/jerboaa/jdk8u-dev/actions/runs/21296331427/job/61304441710#step:12:1032
> 
> The reason is similar to the failures we saw in JDK 11u. Therefore, I propose to backport the GHA work-around to JDK 8u as well. This should stabilize testing.
> 
> Test run with this is here:
> https://github.com/jerboaa/jdk8u-dev/actions/runs/21300159612

Patch is not clean because GHA workflows are setup differently in 8u to 11u. So the same change to a different file is basically it.

-------------

PR Comment: https://git.openjdk.org/jdk8u-dev/pull/749#issuecomment-3792313828