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This repository has been archived by the owner on Oct 14, 2024.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

Promote AdoptOpenJDK Version 8u222 (GA), 11.0.4 (GA), 12.0.2 (GA) #99

Closed
55 of 57 tasks
karianna opened this issue Jul 16, 2019 · 12 comments
Closed
55 of 57 tasks

Promote AdoptOpenJDK Version 8u222 (GA), 11.0.4 (GA), 12.0.2 (GA) #99

karianna opened this issue Jul 16, 2019 · 12 comments
Assignees
Milestone

Comments

@karianna
Copy link
Member

karianna commented Jul 16, 2019

Release Order

We will release the following versions/platforms in the order listed below although this may be subject to change on the day of the release.

Please see the #release channel in the adoptopenjdk.net/slack for coordination efforts.

jdk8u222-b10 - Phase 1

Hotspot JVM

  • x64Linux
  • x64Mac
  • x64Windows

OpenJ9 JVM

  • x64Linux
  • x64Mac
  • x64Windows

11.0.4+11 - Phase 1

Hotspot JVM

  • x64Linux
  • x64Mac
  • x64Windows

OpenJ9 JVM

  • x64Linux
  • x64Mac
  • x64Windows

12.0.2+10 - Phase 1

Hotspot JVM

  • x64Linux
  • x64Mac
  • x64Windows

OpenJ9 JVM

  • x64Linux
  • x64Mac
  • x64Windows

jdk8u222-b10 - Phase 2

Hotspot JVM

  • x32Windows
  • ppc64Aix
  • ppc64leLinux
  • s390xLinux
  • sparcv9Solaris
  • x64Solaris
  • arm32Linux

Hotspot JVM

  • aarch64Linux

OpenJ9 JVM:

  • x32Windows
  • ppc64Aix
  • ppc64leLinux
  • 390xLinux
  • linuxXL
  • x64MacXL
  • x64WindowsXL

11.0.4+11 - Phase 2

Hotspot JVM

  • x32Windows
  • ppc64Aix
  • ppc64leLinux
  • s390xLinux
  • aarch64Linux
  • arm32Linux

OpenJ9 JVM:

  • ppc64Aix
  • ppc64leLinux
  • 390xLinux
  • linuxXL
  • x64MacXL
  • x64WindowsXL

12.0.2+10 - Phase 2

Hotspot JVM

  • x32Windows
  • ppc64Aix
  • ppc64leLinux
  • s390xLinux
  • aarch64Linux
  • arm32Linux

OpenJ9 JVM:

  • ppc64Aix
  • ppc64leLinux
  • 390xLinux
  • linuxXL
  • x64MacXL
  • x64WindowsXL
@karianna karianna added this to the July 2019 milestone Jul 16, 2019
@karianna karianna self-assigned this Jul 16, 2019
@jerboaa
Copy link

jerboaa commented Jul 16, 2019

JDK 8 GA: 8u222-b10
JDK 11 GA: 11.0.4+11
JDK 12 GA: 12.0.2+10

@karianna karianna changed the title Promote AdoptOpenJDK Version 8u222 (GA), 11.0.4 (GA), 12.0.2 (GA), 13.0.0 (GA) Promote AdoptOpenJDK Version 8u222 (GA), 11.0.4 (GA), 12.0.2 (GA) Jul 17, 2019
@karianna
Copy link
Member Author

13 isn't out until Sept - so removed those.

@karianna
Copy link
Member Author

karianna commented Jul 17, 2019

PostMortem

  • Can we speed up tests by a combination of splitting test suites across hosts or simply getting more powerful hardware?
  • Need to think about a completely locked down env to test any embargoed patches so we don't just see them for the first time when the GA tag comes out
  • Can we globally disable all cron jobs on Jenkins?
  • Stabilize machine layer before the day of release (macs and windows machines added JIT, lack soaking period to see how they perform)
  • More AIX and 'lesser' platform hosts required else the whole pipeline gets held up (from a user experience perspective, we could run 1 job pipelines which would speed releases of individual platforms up - but it's easier / better to do the whole lot in one.
  • Some test jobs failed due to timeouts which error the agent. This hangs the job which hangs the pipeline.
  • Need far more robust hosts on the secondary platforms - end users will need to contribute else we should consider dropping support for them
  • Add Homebrew and Docker to the release pipelines (related Third party container tests job should be triggered by docker image update? adoptium/aqa-tests#773, Accommodate other container base images in testing adoptium/aqa-tests#1109 and https://github.com/AdoptOpenJDK/openjdk-build/issues/1123)

@smlambert
Copy link

smlambert commented Jul 17, 2019

Test Analysis Hotspot JVM - jdk8

x64Linux

  • sanity.openjdk - pass
  • sanity.system - 1 failure (TestJlmRemoteMemoryNoAuth_SE80_Linux reproducible in grinder), 10x grinder - 9/10 iterations passing, will mark as non-blocking
  • extended.system - pass
  • sanity.perf - pass
  • sanity.external - 1 failure (tomee microprofile tcks - non-blocking, first fail 3 days ago, will investigate)

x64Mac

  • sanity.openjdk - 2 failures (likely machine issue, ports in use, pass on non-newly added mac in this grinder)
  • sanity.system - 1 failure (TestJlmRemoteMemoryNoAuth_SE80 - grinder - passed on test-macincloud-macos1013-x64-1)
  • extended.system - 1 failure (HCRLateAttachWorkload grinder) - not reproducible in 2nd grinder running on test-macincloud-macos1010-2 - either intermittent or machine issue, will investigate further but marking as not blocking

x64Windows

  • sanity.openjdk - 1 failure (jdk_util: java/util/Locale/LocaleProviders.sh) - reproduced in grinder on test-azure-win2012r2-x64-1, reran on internal machine without failure.
  • sanity.system - passes
  • extended.system - passes

x32Windows

  • sanity.openjdk - 1 failure (jdk_util: java/util/Locale/LocaleProviders.sh same as x64)

ppc64leLinux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

s390xLinux

ppc64Aix

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

aarch64Linux

  • sanity.openjdk - 2 failures - timeouts (need to be calibrated, outrageously long factor)
  • sanity.system - pass
  • extended.system - pass

@smlambert
Copy link

smlambert commented Jul 17, 2019

Test Analysis Hotspot JVM - jdk11

x64Linux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass
  • special.functional - pass
  • sanity.perf - pass
  • sanity.external - pass

x64Mac

  • sanity.openjdk - 6 failures will rerun on separate machine to see if recur
  • 2 of the 6 failures (in jdk_lang and jdk_util targets) fail in similar way (also seen in jdk8 failure)
    java.io.IOException: well-known file /var/folders/h1/b_v48zpd1zd54jblxnx58p3r0000gr/T/.java_pid73679 is not secure: file's group should be the current group (which is -1) but the group is 0
  • sanity.system - pass
  • extended.system - pass
  • sanity.perf - expected compile fail (non-blocking, job updated, rerun & passes here

x64Windows

  • sanity.openjdk - 1 failure (socket in use, will retry on different machine, grinder - non-blocking)
  • sanity.system - pass
  • extended.system - pass

x32Windows

ppc64Aix

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

aarch64Linux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

ppc64leLinux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass
  • sanity.perf - pass

@karianna
Copy link
Member Author

karianna commented Jul 17, 2019

Test Analysis Hotspot JVM - jdk11

After chatting to @smlambert - I'm comfortable with the failing tests for Mac OS X, Windows and Linux x86 with regards to the release quality and will publish.

@karianna
Copy link
Member Author

karianna commented Jul 17, 2019

Test Analysis Hotspot JVM - jdk8

After chatting to @smlambert - I'm comfortable with the failing tests for all of the platforms with regards to the release quality and will publish.

@smlambert
Copy link

smlambert commented Jul 18, 2019

Test Analysis OpenJ9 JVM - jdk8

x64Linux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass
  • sanity.perf - pass
  • sanity.external - 2 failures (tomee microprofile tck same as hotspot, thorntail-mp-tck) - marking this as non-blocking (but we need to triage and report both)
  • special.functional - pass

x64Linux_XL

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

ppc64Aix - sick machine build-osuosl-ppc64-aix-71-1 - will run at ci.eclipse.org/openj9 while its being fixed

  • sanity.openjdk - grinder passes
  • sanity.system
  • extended.system

x64Mac

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - 1 failure (SharedClassesAPI_0 - not included in HS set, passing at ci.eclipse.org/openj9)

x64Mac_XL

x64Windows

  • sanity.openjdk - 1 failure (on some machines, java/util/Locale/LocaleProviders.sh same as hotspot), non-blocking
  • sanity.system
  • extended.system

x32Windows

  • sanity.openjdk - 1 failure same as Hotspot (non-blocking)

ppc64leLinux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

s390xLinux

  • sanity.openjdk - 1 failure (non-blocking, java/nio/channels/Selector/LotsOfChannels.java)
  • sanity.system - 8 failures - all JLMRemoteMemory/Class tests with connection issues (will rerun on different machine, passes at ci.eclipse.org/openj9) - grinder rerun passes
  • extended.system - pass

@smlambert
Copy link

smlambert commented Jul 18, 2019

Test Analysis OpenJ9 JVM - jdk11

x64LInux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass
  • sanity.perf - pass
  • sanity.external - pass
  • special.functional - pass

x64LInux_XL

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

x64Mac

  • sanity.openjdk - 2 failures - non-blocking
  • sanity.system - pass
  • extended.system - 2 failures (SharedClassesAPI grinder-1, -2, SharedClasses.SCM23.MultiThread grinder-1, -2 both pass) - passes in runs at ci.eclipse.org/openj9

x64Mac_XL

  • sanity.openjdk - 5 failures - non-blocking (grinder-1) passes (run on test-macincloud-macos1014-x64-2)
  • sanity.system - pass
  • extended.system - 1 failure - same as jdk8, SharedClassesAPI_0 (passing at ci.eclipse.org/openj9)

x64Windows

  • sanity.openjdk - 1 failure - socket in use, non-blocking (passes on different machine, internal Grinder_Advanced/334)
  • sanity.system - pass
  • extended.system - pass

x64Windows_XL

  • sanity.openjdk - 2 failures - BitSetStreamTest grinder passes, non-blocking
  • sanity.system - pass

s390xLinux

  • sanity.openjdk - 7 failures - connection issues / timeouts
  • sanity.system - still running (expect to see connection issues on -XJ machine, passes at ci.eclipse.org/openj9)
  • extended.system - pass

ppc64Aix - aix test machine down, will run at ci.eclipse.org/openj9

  • sanity.openjdk - grinder run
  • sanity.system
  • extended.system

ppc64leLinux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

@smlambert
Copy link

smlambert commented Jul 18, 2019

Test Analysis Hotspot JVM - jdk12

x86Windows

  • sanity.openjdk - 1 failure - non-blocking (java/net/DatagramSocket/ReuseAddressTest.java address in use)
  • sanity.system - pass
  • extended.system - pass

x86Linux

  • sanity.openjdk - pass
  • sanity.system - 1 failure - TestJlmRemoteMemoryAuth_0, non-blocking same as jdk8
  • extended.system - pass

x86Mac

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - 1 failure - HCRLateAttachWorkload_0 (same as jdk8, non-blocking)

x32Windows

ppcle64Linux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

s390xLinux

  • sanity.openjdk
  • sanity.system - pass
  • extended.system - pass

aarch64Linux

  • sanity.openjdk - pass
  • sanity.system - 3 failures need investigation (MauveSingleThreadLoadTest_HotSpot, MauveSingleInvocationLoadTest_HotSpot, MauveMultiThreadLoadTest_HS)
  • extended.system - 2 failures that need investigation (MathLoadTest, LockingLoadTest)

Waiting on good builds for arm32 and ppc64Aix

@smlambert
Copy link

smlambert commented Jul 19, 2019

Test Analysis OpenJ9 JVM - jdk12

x64Windows

  • sanity.openjdk - 1 failure - non-blocking (ReuseAddressTest)
  • sanity.system - pass
  • extended.system - pass

x64WindowsXL

s390xLinux

  • sanity.openjdk - 6 failures - jdk_net, non-blocking
  • sanity.system - pass
  • extended.system - pass

x64Mac

  • sanity.openjdk - 3 failures (comm. errors seen before, non-blocking)
  • sanity.system - pass
  • extended.system - 1 failure (SharedClassesAPI, non-blocking)

x64MacXL

  • sanity.openjdk - 1 failure (non-blocking)
  • sanity.system - pass
  • extended.system - 1 failure (SharedClassesAPI, non-blocking)

x64Linux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - 1 failure (SharedClassesAPI, non-blocking)

x64LinuxXL

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

ppc64leLinux

  • sanity.openjdk - pass
  • sanity.system - pass
  • extended.system - pass

ppc64Aix build failed for a known reason
SharedClassesAPI failure is due to a machine config issue (memory mapped shared memory should get increased)

@dinogun
Copy link

dinogun commented Jul 19, 2019

Prereq builds needed for updating the Official Docker Images.

jdk8u222-b10

Hotspot JVM

  • x64Linux
  • ppc64leLinux
  • s390xLinux
  • x64Windows
  • arm32Linux

OpenJ9 JVM

  • ppc64leLinux
  • s390xLinux
  • x64Linux
  • x64Windows

11.0.4+11

Hotspot JVM

  • x64Linux
  • ppc64leLinux
  • s390xLinux
  • aarch64Linux
  • x64Windows
  • arm32Linux

OpenJ9 JVM

  • ppc64leLinux
  • s390xLinux
  • x64Linux
  • x64Windows

12.0.2+10

Hotspot JVM

  • x64Linux
  • ppc64leLinux
  • s390xLinux
  • aarch64Linux
  • x64Windows
  • arm32Linux

OpenJ9 JVM

  • ppc64leLinux
  • s390xLinux
  • x64Linux
  • x64Windows

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

5 participants