Hi, I think, if we can't reproduce with later versions, we wont need to open bug report. This happened on ZGC (which is experimental on JDK 12). I know from my talks last FOSDEM and the many open issues we have seen at that time, that it may have been fxed. Nowadays, due to randomization, I know that JDK 12 in combination with ZGC is running quite seldom.
I would at some point cleanup policeman jenkins and remove the short-term versions (12, 13, 14) and only keep the latest short term and all LTS releases available. This would also raise the probability that the randomization chooses a JVM/GC combination which is "uptodate" and "often used" (LTS). What do you think? Uwe ----- Uwe Schindler Achterdiek 19, D-28357 Bremen https://www.thetaphi.de eMail: [email protected] > -----Original Message----- > From: Dawid Weiss <[email protected]> > Sent: Friday, December 11, 2020 9:08 AM > To: Lucene Dev <[email protected]> > Cc: [email protected] > Subject: Re: [JENKINS-Experimental-GC] Lucene-Solr-master-Linux (64bit/jdk- > 12.0.2) - Build # 28914 - Still Failing! > > It's jdk 12.0.2+10... do you think there will be interest in tracking > this?... I have one that popped up on jdk14+ recently and it's quite > terrifying as it doesn't result in a crash - it's wrong runtime > behavior [1]... I think I'll focus on this one first. > > D. > > [1] https://mail.openjdk.java.net/pipermail/hotspot-dev/2020- > December/047240.html > > On Fri, Dec 11, 2020 at 4:08 AM Robert Muir <[email protected]> wrote: > > > > Should we report the bug? Worst case it is a duplicate? > > > > SIGILL makes it clear it's a compiler issue. maybe they have a way to > > quickly decode the stuff from that "instructions" list in hs_err.log > > and figure out what happened (e.g. unavailable avx instruction used or > > whatever went wrong)? > > > > I attached the log, so it isn't lost from jenkins builds. > > > > On Wed, Dec 9, 2020 at 3:52 AM Dawid Weiss <[email protected]> > wrote: > > > > > > These seem to be JVM errors (in test runners). > > > > > > # > > > # A fatal error has been detected by the Java Runtime Environment: > > > # > > > # SIGILL (0x4) at pc=0x00007f10f70cee26, pid=450437, tid=557573 > > > # > > > # JRE version: OpenJDK Runtime Environment (12.0.2+10) (build 12.0.2+10) > > > # Java VM: OpenJDK 64-Bit Server VM (12.0.2+10, mixed mode, tiered, z > > > gc, linux-amd64) > > > # Problematic frame: > > > # J 47230 c2 > org.apache.solr.schema.TrieField.readableToIndexed(Ljava/lang/CharSequence; > Lorg/apache/lucene/util/BytesRefBuilder;)V > > > (147 bytes) @ 0x00007f10f70cee26 > > > [0x00007f10f70cea60+0x00000000000003c6] > > > # > > > # No core dump will be written. Core dumps have been disabled. To > > > enable core dumping, try "ulimit -c unlimited" before starting Java > > > again > > > # > > > # An error report file with more information is saved as: > > > # /home/jenkins/workspace/Lucene-Solr-master- > Linux/solr/core/build/tmp/tests-cwd/hs_err_pid450437.log > > > > > > On Wed, Dec 9, 2020 at 7:20 AM Policeman Jenkins Server > > > <[email protected]> wrote: > > > > > > > > Build: https://jenkins.thetaphi.de/job/Lucene-Solr-master-Linux/28914/ > > > > Java: 64bit/jdk-12.0.2 -XX:-UseCompressedOops - > XX:+UnlockExperimentalVMOptions -XX:+UseZGC > > > > > > > > All tests passed > > > > > > > > --------------------------------------------------------------------- > > > > To unsubscribe, e-mail: [email protected] > > > > For additional commands, e-mail: [email protected] > > > > > > --------------------------------------------------------------------- > > > To unsubscribe, e-mail: [email protected] > > > For additional commands, e-mail: [email protected] > > > > > > > --------------------------------------------------------------------- > > To unsubscribe, e-mail: [email protected] > > For additional commands, e-mail: [email protected] > > --------------------------------------------------------------------- > To unsubscribe, e-mail: [email protected] > For additional commands, e-mail: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
