Skip to content
This repository has been archived by the owner on Oct 14, 2024. It is now read-only.

Promote AdoptOpenJDK Version jdk-13.0.2+8 #134

Closed
2 tasks done
andrew-m-leonard opened this issue Jan 17, 2020 · 24 comments
Closed
2 tasks done

Promote AdoptOpenJDK Version jdk-13.0.2+8 #134

andrew-m-leonard opened this issue Jan 17, 2020 · 24 comments

Comments

@andrew-m-leonard
Copy link

andrew-m-leonard commented Jan 17, 2020

Java Version:

JVM:

  • OpenJ9
  • Hotspot

@AdoptOpenJDK/tsc please can somebody +1 this request?

@andrew-m-leonard
Copy link
Author

jdk-13.0.2+8_openj9-0.18.0 submitted, 2 jobs queued:
All platforms except AIX: https://ci.adoptopenjdk.net/job/build-scripts/job/openjdk13-pipeline/171/
AIX: https://ci.adoptopenjdk.net/job/build-scripts/job/openjdk13-pipeline/172/

@andrew-m-leonard
Copy link
Author

AIX job aborted, to allow Hotspot pipeline to run

@andrew-m-leonard
Copy link
Author

andrew-m-leonard commented Jan 17, 2020

jdk-13.0.2+8_openj9-0.18.0 :
https://ci.adoptopenjdk.net/job/build-scripts/job/openjdk13-pipeline/171/

All these platforms tests all Green, no failures:

  • pLinux
  • pLinuxXL
  • zLinuxXL
  • xLinux
  • xLinuxXL
  • windows
  • windowsXL

Build failures for these platforms, needing re-build:

  • zLinux
  • macOS
  • macOSXL

Not built yet:

  • AIX

@smlambert
Copy link

Great to see all 751 test targets passing! Reminder that each test target contains many tests, often thousands!

Screen Shot 2020-01-17 at 11 21 20 AM

@karianna karianna added this to the January 2020 milestone Jan 18, 2020
@andrew-m-leonard
Copy link
Author

andrew-m-leonard commented Jan 18, 2020

jdk-13.0.2+8_openj9-0.18.0 :

@gdams
Copy link
Member

gdams commented Jan 20, 2020

@gdams
Copy link
Member

gdams commented Jan 20, 2020

jdk-13.0.2+8 Failures:
Mac
extended.system
HCRLateAttachWorkload_0 => Failed on 4/5 last builds. Raised openjdk-tests/1555 - looks like a long-standing, recurring issue (as seen in deep history view)

Linux aarch64
sanity.system => non blockers (see #134 (comment))
MauveMultiThreadLoadTest_0
MauveSingleInvocationLoadTest_HotSpot_0
MauveSingleThreadLoadTest_HotSpot_0

sanity.openjdk => All intermittently fail on upstream too (see #132 (comment))
java/lang/invoke/VarHandles/VarHandleTestAccessFloat.java
java/lang/invoke/VarHandles/VarHandleTestByteArrayAsInt.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessString.java
java/lang/invoke/VarHandles/VarHandleTestAccessChar.java
java/lang/invoke/VarHandles/VarHandleTestAccessDouble.java
java/lang/invoke/VarHandles/VarHandleTestAccessInt.java
java/lang/invoke/VarHandles/VarHandleTestAccessBoolean.java
java/lang/invoke/VarHandles/VarHandleTestAccessByte.java
java/lang/invoke/VarHandles/VarHandleTestAccessLong.java
java/lang/invoke/VarHandles/VarHandleTestAccessShort.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessBoolean.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessByte.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessChar.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessDouble.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessFloat.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessLong.java
java/lang/invoke/VarHandles/VarHandleTestMethodHandleAccessShort.java

extended.system
MathLoadTest_all_0 => been failing long term deep history
LockingLoadTest_0 => been failing long term deep history

@andrew-m-leonard
Copy link
Author

jdk-13.0.2+8_openj9-0.18.0 :

zLinux, mac, macXL builds & tests clean, ready for publish

AIX built successfully, but tests unable to run:

20:56:40  libjvm.so preloadLibrary(/ramdisk0/build/workspace/Test_openjdk13_j9_sanity.system_ppc64_aix/openjdkbinary/j2sdk-image/lib/compressedrefs/libj9vm29.so): 	0509-022 Cannot load module /ramdisk0/build/workspace/Test_openjdk13_j9_sanity.system_ppc64_aix/openjdkbinary/j2sdk-image/lib/compressedrefs/libj9vm29.so.
20:56:40  	0509-150   Dependent module libc++.a(cxxabi_64.o) could not be loaded.
20:56:40  	0509-022 Cannot load module libc++.a(cxxabi_64.o).
20:56:40  	0509-026 System error: A file or directory in the path name does not exist.

@gdams
Copy link
Member

gdams commented Jan 20, 2020

@andrew-m-leonard +1 to shipping all clean platforms

@andrew-m-leonard
Copy link
Author

@gdams I was thinking that, however publishing the pipeline is going to publish AIX as it built but can;t be tested.....

@gdams
Copy link
Member

gdams commented Jan 20, 2020

@gdams I was thinking that, however publishing the pipeline is going to publish AIX as it built but can;t be tested.....

@smlambert do you know of a way to work around this?

@andrew-m-leonard
Copy link
Author

andrew-m-leonard commented Jan 20, 2020

@gdams Stewart @sxa555 suggested re-launching failed test jobs on node build-ibm-ppc64-aix-71-1
this seems to be working so far
If sanity.openjdk is good, I suggest we publish

@smlambert
Copy link

+1 to publishing.

@gdams
Copy link
Member

gdams commented Jan 20, 2020

phase 1 publishing complete.

@smlambert
Copy link

smlambert commented Jan 20, 2020

re: #134 (comment), I will launch Grinders to retest the aarch64 system test failures.

Looking at the mauve failures, they have been failing for quite some time (do not see that they ever passed on aarch64). For release retrospective, I would like to see more regular triaging of hotspot nightly tests, as we have seen at least 4 test targets that have been failing for weeks, that could have been triaged prior to the release.

MauveSingleInvocationLoadTest_HotSpot_0:
LT 02:23:30.549 - Ran : 3850
LT 02:23:30.549 - Passed : 3848
LT 02:23:30.549 - Failed : 2

MauveMultiThreadLoadTest_0:
LT 02:23:37.516 - Ran : 9038
LT 02:23:37.516 - Passed : 9028
LT 02:23:37.516 - Failed : 10

MauveSingleThreadLoadTest_HotSpot_0:
LT 02:23:21.817 - Ran : 17394
LT 02:23:21.817 - Passed : 17384
LT 02:23:21.818 - Failed : 10

All failing on same subtest at some point in the repetitive build up of load in these test runs:
Test details = 'Mauve[gnu.testlet.java.lang.Math.strictfp_modifier.cos]'

@andrew-m-leonard
Copy link
Author

jdk-13.0.2+8_openj9-0.18.0 on AIX sanity.openjdk failures:
java/lang/ProcessBuilder/Basic.java#id0.Basic_id0
java/lang/invoke/condy/CondyNestedResolutionTest.java.CondyNestedResolutionTest
java/net/Inet4Address/PingThis.java.PingThis
java/net/Socket/asyncClose/Race.java.Race
java/net/SocketPermission/SocketPermissionTest.java.SocketPermissionTest
java/nio/channels/AsyncCloseAndInterrupt.java.AsyncCloseAndInterrupt
java/util/concurrent/locks/Lock/TimedAcquireLeak.java.TimedAcquireLeak
java/util/logging/LogManagerAppContextDeadlock.java.LogManagerAppContextDeadlock

@andrew-m-leonard
Copy link
Author

@smlambert Are you ok to published with the above failures on AIX?

@smlambert
Copy link

Given that the testing for the JDK13 AIX builds have moved onto the AIX71TL05SP5 machine, and we have seen these jdk_lang networking failures previously on machines newly setup for testing AND we run a deeper set of sanity.functional targets that exercise this same functionality on openj9 builds, I think we are good to proceed to publish AIX.

@andrew-m-leonard
Copy link
Author

Thanks @smlambert i'll go ahead and publish

@andrew-m-leonard
Copy link
Author

jdk-13.0.2+8_openj9-0.18.0 all platforms published.

@gdams
Copy link
Member

gdams commented Jan 21, 2020

shipping phase 2 platforms for hotspot (except for AIX which needs a rebuild) as triaging is complete.

@gdams
Copy link
Member

gdams commented Jan 21, 2020

just AIX to go. The builds have been failing regularly (@sxa555 any idea what is going on here? https://ci.adoptopenjdk.net/job/build-scripts/job/jobs/job/jdk13u/job/jdk13u-aix-ppc64-hotspot/156/console)

@M-Davies
Copy link

@gdams I've seen that failure before on AIX adoptium/infrastructure#1099

@karianna karianna modified the milestones: January 2020, February 2020 Feb 12, 2020
@sxa
Copy link
Member

sxa commented Apr 3, 2020

These are all out now AFAIK therefore closing

Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Development

No branches or pull requests

6 participants