It happened at the same CompileTrask on our land (OpenJDK17):

Thanks,
Jen-Ya
---------------  T H R E A D  ---------------

Current thread (0x00007ff2f4a44630):  JavaThread "C2 CompilerThread0"
daemon [_thread_in_native, id=14826,
stack(0x00007ff2d0429000,0x00007ff2d052a000)]


Current CompileTask:
C2:80302242 17311   !   4
com.github.benmanes.caffeine.cache.BoundedLocalCache::put (731 bytes)

Stack: [0x00007ff2d0429000,0x00007ff2d052a000],  sp=0x00007ff2d0524d70,
 free space=1007k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native
code)
V  [libjvm.so+0xad1332]  PhaseIdealLoop::build_loop_late_post_work(Node*,
bool)+0xf2
V  [libjvm.so+0xad1d80]  PhaseIdealLoop::build_loop_late(VectorSet&,
Node_List&, Node_Stack&)+0x180
V  [libjvm.so+0xad26de]  PhaseIdealLoop::build_and_optimize()+0x78e
V  [libjvm.so+0x5d06c1]  PhaseIdealLoop::optimize(PhaseIterGVN&,
LoopOptsMode)+0x171
V  [libjvm.so+0x5ce5b5]  Compile::Optimize()+0xd85
V  [libjvm.so+0x5cff09]  Compile::Compile(ciEnv*, ciMethod*, int, bool,
bool, bool, bool, bool, DirectiveSet*)+0xe49
V  [libjvm.so+0x50f7d7]  C2Compiler::compile_method(ciEnv*, ciMethod*, int,
bool, DirectiveSet*)+0xf7
V  [libjvm.so+0x5d8f59]
 CompileBroker::invoke_compiler_on_method(CompileTask*)+0xf69
V  [libjvm.so+0x5d9c00]  CompileBroker::compiler_thread_loop()+0x4f0
V  [libjvm.so+0xe607b2]  JavaThread::thread_main_inner()+0x182
V  [libjvm.so+0xe63f7e]  Thread::call_run()+0xde
V  [libjvm.so+0xc1b361]  thread_native_entry(Thread*)+0xe1

On Thu, Oct 13, 2022 at 12:28 PM Wei <[email protected]> wrote:

> Looks our case is triggered from caffeine cache, different from
> https://bugs.openjdk.org/browse/JDK-8285835 where it happens in facet
> module.
>
>
> ---------------  T H R E A D  ---------------
>
>
> Current thread (0x00007f96b1286f40):  JavaThread "C2 CompilerThread0"
> daemon [_thread_in_native, id=16637,
> stack(0x00007f85f50ab000,0x00007f85f51ac000)]
>
>
>
> Current CompileTask:
>
> C2:148683516 28882   !   4
> com.github.benmanes.caffeine.cache.BoundedLocalCache::put
> (731 bytes)
>
>
> Stack: [0x00007f85f50ab000,0x00007f85f51ac000],  sp=0x00007f85f51a6e40,
> free
> space=1007k
>
> Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native
> code)
>
> V  [libjvm.so+0x9e13a1]  PhaseIdealLoop::build_loop_late_post_work(Node*,
> bool)+0x111
>
> V  [libjvm.so+0x9e1910]  PhaseIdealLoop::build_loop_late(VectorSet&,
> Node_List&, Node_Stack&)+0x180
>
> V  [libjvm.so+0x9e2321]
> PhaseIdealLoop::build_and_optimize(LoopOptsMode)+0x841
>
> V  [libjvm.so+0x5b90da]  PhaseIdealLoop::optimize(PhaseIterGVN&,
> LoopOptsMode)+0x17a
>
> V  [libjvm.so+0x5b6b5e]  Compile::Optimize()+0x84e
>
> V  [libjvm.so+0x5b883c]  Compile::Compile(ciEnv*, ciMethod*, int, bool,
> bool, bool, bool, bool, DirectiveSet*)+0x13ec
>
> V  [libjvm.so+0x4f87da]  C2Compiler::compile_method(ciEnv*, ciMethod*, int,
> bool, DirectiveSet*)+0xba
>
> V  [libjvm.so+0x5c0c8d]
> CompileBroker::invoke_compiler_on_method(CompileTask*)+0xf8d
>
> V  [libjvm.so+0x5c15d8]  CompileBroker::compiler_thread_loop()+0x588
>
> V  [libjvm.so+0xde6bb8]  JavaThread::thread_main_inner()+0x158
>
> V  [libjvm.so+0xde6cf9]  JavaThread::run()+0x129
>
> V  [libjvm.so+0xde9bce]  Thread::call_run()+0x7e
>
> V  [libjvm.so+0xb337b1]  thread_native_entry(Thread*)+0xe1
>
> On Thu, Oct 13, 2022 at 12:00 PM Wei <[email protected]> wrote:
>
> > We also observed the same SIGSEGV crash during our test with Solr 9.0.0.
> > It happens very rarely though,  seem 2 occurences during one month of
> > testing.  We use Zulu JDK 17.34.
> >
> >
> >    -
> >
> >    A fatal error has been detected by the Java Runtime Environment:
> >    SIGSEGV (0xb) at pc=0x00007f96b852d3a1, pid=16597, tid=16637
> >    # JRE version: OpenJDK Runtime Environment Zulu17.34+20-SA (17.0.3+7)
> >    (build 17.0.3+7-LTS)
> >    # Java VM: OpenJDK 64-Bit Server VM Zulu17.34+20-SA (17.0.3+7-LTS,
> >    mixed mode, sharing, tiered, compressed class ptrs, g1 gc,
> linux-amd64)
> >    # Problematic frame:
> >    # V  [libjvm.so+0x9e13a1]
> >    PhaseIdealLoop::build_loop_late_post_work(Node*, bool)+0x111
> >
> >
> > Do we know which lucene/solr component would trigger this error?
> >
> > Regards,
> > Wei
> >
> > On Thu, Oct 13, 2022 at 10:51 AM solr <[email protected]> wrote:
> >
> >> Hi.
> >>
> >> We tried to upgrade a number of solar images from 8.11 to 9.0.0 this
> week
> >> and ran into the same problems with the official docker image as well
> >> (approximately 30 running pods in Kubernetes, crashing on average once
> an
> >> hour).
> >>
> >> We even tried to build a docker image with another base image
> >> (bellsoft/liberica-openjdk-debian:17), with the same result (even though
> >> the actual error differed somewhat (optimize instead of
> >> build_loop_late_post_work)).
> >>
> >> For now we ended up with building a docker image of solr 9.0.0 based on
> >> openjdk:11-jre and it has been running several hours without crashes.
> >> Maybe the official docker image should be reverted to this until the
> bug is
> >> fixed?
> >>
> >>
> >> For reference: here are 2 crashes (hs_err_pidXX.log and replay_pidXX.log
> >> can be provided on request):
> >>
> >>
> >> Temurin-17.0.4.1+1:
> >> #
> >> # A fatal error has been detected by the Java Runtime Environment:
> >> #
> >> #  SIGSEGV (0xb) at pc=0x00007fd81c7e6153, pid=14, tid=85
> >> #
> >> # JRE version: OpenJDK Runtime Environment Temurin-17.0.4.1+1
> >> (17.0.4.1+1) (build 17.0.4.1+1)
> >> # Java VM: OpenJDK 64-Bit Server VM Temurin-17.0.4.1+1 (17.0.4.1+1,
> mixed
> >> mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc,
> >> linux-amd64)
> >> # Problematic frame:
> >> # V  [libjvm.so+0xacc153]
> >> PhaseIdealLoop::build_loop_late_post_work(Node*, bool)+0x153
> >> #
> >> # Core dump will be written. Default location: /core.%e.14.%t
> >> #
> >> # An error report file with more information is saved as:
> >> # /opt/solr-9.0.0/server/hs_err_pid14.log
> >> {#
> >> # Compiler replay data is saved as:
> >> # /opt/solr-9.0.0/server/replay_pid14.log
> >> #
> >> # If you would like to submit a bug report, please visit:
> >> #   https://github.com/adoptium/adoptium-support/issues
> >> #
> >> /opt/scripts/startSolr.sh: line 54:    14 Aborted
> >>  /opt/solr/bin/solr -f -p ${solrport} -m ${memory}
> >> -Dsolr.jetty.request.header.size=65535 -Dsolr.disable.allowUrls=true
> >> -Dsolrindex=${profile} -Dlog4j2.formatMsgNoLookups=true
> >> -Dfinnapp=${finnapp} -Dsolr.master.host=${masterhost}
> -Denable.slave=true
> >> -Denable.replication=true -Dsolr.http1=true
> >> -Dsolr.disable.shardsWhitelist=true -Dodin.job.host=solr-cl-job
> >> -Dodin.bap.host=solr-cl-bap -Dodin.motor.host=solr-cl-motor
> >> -Dodin.estate.host=solr-cl-estate
> >>
> >>
> >> bellsoft/liberica-openjdk-debian:17
> >> #
> >> # A fatal error has been detected by the Java Runtime Environment:
> >> #
> >> #  SIGSEGV (0xb) at pc=0x00007f96e2515e7f, pid=13, tid=76
> >> #
> >> # JRE version: OpenJDK Runtime Environment (17.0.4.1+1) (build
> >> 17.0.4.1+1-LTS)
> >> # Java VM: OpenJDK 64-Bit Server VM (17.0.4.1+1-LTS, mixed mode, tiered,
> >> compressed oops, compressed class ptrs, g1 gc, linux-amd64)
> >> # Problematic frame:
> >> # V  [libjvm.so+0x515e7f]  PhaseIdealLoop::optimize(PhaseIterGVN&,
> >> LoopOptsMode)+0x13bf
> >> #
> >> # Core dump will be written. Default location: /core.%e.13.%t
> >> #
> >> # An error report file with more information is saved as:
> >> # /opt/solr-9.0.0/server/hs_err_pid13.log
> >> #
> >> # Compiler replay data is saved as:
> >> # /opt/solr-9.0.0/server/replay_pid13.log
> >> #
> >> # If you would like to submit a bug report, please visit:
> >> #   https://bell-sw.com/support
> >> #
> >> /opt/scripts/startSolr.sh: line 54:    13 Aborted
> >>  /opt/solr/bin/solr -f -p ${solrport} -m ${memory}
> >> -Dsolr.jetty.request.header.size=65535 -Dsolr.disable.allowUrls=true
> >> -Dsolrindex=${profile} -Dlog4j2.formatMsgNoLookups=true
> >> -Dfinnapp=${finnapp} -Dsolr.master.host=${masterhost}
> -Denable.slave=true
> >> -Denable.replication=true -Dsolr.http1=true
> >> -Dsolr.disable.shardsWhitelist=true -Dodin.job.host=solr-cl-job
> >> -Dodin.bap.host=solr-cl-bap -Dodin.motor.host=solr-cl-motor
> >> -Dodin.estate.host=solr-cl-estate
> >>
> >>
> >>
> >>
> >>
> >>
> >> Regards,
> >>
> >>
> >> Fredrik
> >>
> >>
> >> --
> >>
> >> Fredrik Rødland               Cell:    +47 99 21 98 17
> >> Maisen Pedersens vei 1        Twitter: @fredrikr
> >> NO-1363 Høvik, NORWAY         flickr:
> >> http://www.flickr.com/fmmr/
> >> http://rodland.no             about.me http://about.me/fmr
> >>
> >>
>

Reply via email to