8369226: GHA: Switch to MacOS 15

[vieiro]

Backport of JDK-8369226 from JDK17 that upgrades macos-13 runners to macos-15-intel, and XCode from 14.3.1 to 16.4, since macos-13 will be removed in early december.

The backport is not clean, as there are many differences between GHA in JDK11 and upper versions (e.g.
JDK-8341097, JDK-8325194).

---

Progress

• Change must be properly reviewed (1 review required, with at least 1 Reviewer)
• Change must not contain extraneous whitespace
• Commit message must refer to an issue
• JDK-8369226 needs maintainer approval
• JDK-8336451 needs maintainer approval

Issues

• JDK-8369226: GHA: Switch to MacOS 15 (Enhancement - P4)
• JDK-8336451: [11u] GHA macos-13 and macos-15 builders are unable to resolve local hostname (Enhancement - P4)

Reviewing

Using git

Checkout this PR locally:
$ git fetch https://git.openjdk.org/jdk11u-dev.git pull/3117/head:pull/3117
$ git checkout pull/3117

Update a local copy of the PR:
$ git checkout pull/3117
$ git pull https://git.openjdk.org/jdk11u-dev.git pull/3117/head

Using Skara CLI tools

Checkout this PR locally:
$ git pr checkout 3117

View PR using the GUI difftool:
$ git pr show -t 3117

Using diff file

Download this PR as a diff file:
https://git.openjdk.org/jdk11u-dev/pull/3117.diff

Using Webrev

Link to Webrev Comment

[vieiro]

/issue add JDK-8368192

[bridgekeeper]

👋 Welcome back avieiro! A progress list of the required criteria for merging this PR into master will be added to the body of your pull request. There are additional pull request commands available for use with this pull request.

[openjdk]

❗ This change is not yet ready to be integrated.
See the Progress checklist in the description for automated requirements.

[openjdk]

This backport pull request has now been updated with issue from the original commit.

[openjdk]

@vieiro
Adding additional issue to issue list: 8368192: Test java/lang/ProcessBuilder/Basic.java#id0 fails with Exception: Stack trace.

[mlbridge]

Webrevs

• 02: Full - Incremental (2e624f30)
• 01: Full - Incremental (f607e393)
• 00: Full (fefa04dd)

[vieiro]

One JDWP test failed when using a *.local address for JDWP:

-Xrunjdwp:transport=dt_socket,address=Mac-1762967583261.local:49212

With:

2025-11-12T17:27:03.7710510Z ERROR: transport error 202: getaddrinfo: unknown host
2025-11-12T17:27:03.7742510Z ERROR: JDWP Transport dt_socket failed to initialize, TRANSPORT_INIT(510)
2025-11-12T17:27:03.7762640Z JDWP exit error AGENT_ERROR_TRANSPORT_INIT(197): No transports initialized [src/jdk.jdwp.agent/share/native/libjdwp/debugInit.c:745]

This is nothing new. macos-13 GHA runners were also unable to resolve *.local addresses using getaddrinfo (see this issue, or this reproducer ) and the problem may be also happening in macos-15-intel GHA runners. I'll try to reproduce and file a new issue.

[jerboaa]

Please move JDK-8368192 to a separate PR/backport.

[vieiro]

/issue remove JDK-8368192

[openjdk]

@vieiro
Removing additional issue from issue list: 8368192.

[vieiro]

(Dropped commit for JDK-8368192 ).

[jerboaa]

One JDWP test failed when using a *.local address for JDWP:

-Xrunjdwp:transport=dt_socket,address=Mac-1762967583261.local:49212

With:

2025-11-12T17:27:03.7710510Z ERROR: transport error 202: getaddrinfo: unknown host
2025-11-12T17:27:03.7742510Z ERROR: JDWP Transport dt_socket failed to initialize, TRANSPORT_INIT(510)
2025-11-12T17:27:03.7762640Z JDWP exit error AGENT_ERROR_TRANSPORT_INIT(197): No transports initialized [src/jdk.jdwp.agent/share/native/libjdwp/debugInit.c:745]

This is nothing new. macos-13 GHA runners were also unable to resolve *.local addresses using getaddrinfo (see this issue, or this reproducer ) and the problem may be also happening in macos-15-intel GHA runners. I'll try to reproduce and file a new issue.

I think we should consider with the work-around proposed in:
#2861

[openjdk]

@vieiro Please do not rebase or force-push to an active PR as it invalidates existing review comments. Note for future reference, the bots always squash all changes into a single commit automatically as part of the integration. See OpenJDK Developers’ Guide for more information.

[vieiro]

This is nothing new. macos-13 GHA runners were also unable to resolve *.local addresses using getaddrinfo (see this issue, or this reproducer ) and the problem may be also happening in macos-15-intel GHA runners. I'll try to reproduce and file a new issue.

I think we should consider with the work-around proposed in: #2861

Yep. This will shield the tests against resolution of .local addresses in macos.

FWIW the GHA issue on macos-15 was already reported here.

I'll await for the test results and may add the fix in #2861 as a separate commit afterwards.

[vieiro]

So it seems we could surpass the .local resolving issue, the only error I can see now is:

2025-11-13T12:00:54.2012680Z TEST: java/lang/ProcessBuilder/Basic.java#id0
2025-11-13T12:00:54.2278160Z TEST JDK: /Users/runner/work/jdk11u-dev/jdk11u-dev/bundles/jdk/jdk-11.0.30.jdk/Contents/Home
[...]
2025-11-13T12:00:58.3725770Z cmdp 4: Basic$JavaChild
2025-11-13T12:00:58.3798730Z cmdp 5: System.getenv()
2025-11-13T12:00:58.3846970Z envp
2025-11-13T12:00:58.3929260Z envp 0: LC_ALL=C��
2025-11-13T12:00:58.3968070Z envp 1: FO�=B�R
2025-11-13T12:00:58.3974390Z  waitFor process: delta: 57571
2025-11-13T12:00:58.4017550Z 
2025-11-13T12:00:58.5020540Z Passed = 9265, failed = 3
2025-11-13T12:00:58.5274170Z 
2025-11-13T12:00:58.5428570Z STDERR:
2025-11-13T12:00:58.5528960Z >'TMPDIR=/var/folders/vx/zmvzwd094y53mqcpn3jbm7580000gn/T/,'< not equal to ''
2025-11-13T12:00:58.5587170Z java.lang.Exception: Stack trace
2025-11-13T12:00:58.5623280Z 	at java.base/java.lang.Thread.dumpStack(Thread.java:1383)
2025-11-13T12:00:58.5651390Z 	at Basic.fail(Basic.java:2825)
2025-11-13T12:00:58.5685880Z 	at Basic.fail(Basic.java:2826)
2025-11-13T12:00:58.5702940Z 	at Basic.equal(Basic.java:2832)
2025-11-13T12:00:58.5714790Z 	at Basic.realMain(Basic.java:1375)
2025-11-13T12:00:58.5733890Z 	at Basic.main(Basic.java:2835)
[...]
2025-11-13T12:14:49.8278630Z ==============================
2025-11-13T12:14:49.8281100Z Test summary
2025-11-13T12:14:49.8283090Z ==============================
2025-11-13T12:14:49.8285080Z    TEST                                              TOTAL  PASS  FAIL ERROR   
2025-11-13T12:14:49.8287830Z >> jtreg:test/jdk:tier1_part1                          818   817     1     0 <<
2025-11-13T12:14:49.8290620Z ==============================

Which is JDK-8368192. New PR on the way...

[vieiro]

/issue add JDK-8336451

[openjdk]

@vieiro
Adding additional issue to issue list: 8336451: [11u] GHA macos-13 and macos-15 builders are unable to resolve local hostname.

[vieiro]

Let's add a fix for the .local name resolution using getaddrinfo (see actions/runner-images#12562). This was reported in JDK-8336451

These DNS resolution issues seem to be random, and have reappeared again in #3119 . The fix makes macos tests more deterministic.

[jerboaa]

These DNS resolution issues seem to be random, and have reappeared again in #3119 . The fix makes macos tests more deterministic.

Yes. That's the experience I had as well.

[jerboaa]

Please move JDK-8336451 to a separate PR. AFAIK, we see it in current GHA as well (occasionally) and it's not nice to bundle an initial fix (of a bug) with another unless they are actually related. In this case it's better to separate it out since we'd have a better trace which fixes what. Thanks!