Synchrony is killed with SIGSEGV (0xb) - G1ParScanThreadState::copy_to_survivor_space
プラットフォームについて: Data Center のみ。 - This article only applies to Atlassian apps on the Data Center プラットフォーム。
この KB は Data Center バージョンの製品用に作成されています。Data Center 固有ではない機能の Data Center KB は、製品のサーバー バージョンでも動作する可能性はありますが、テストは行われていません。 Server* 製品のサポートは 2024 年 2 月 15 日に終了しました。Server 製品を実行している場合は、 アトラシアン Server サポート終了 のお知らせにアクセスして、移行オプションを確認してください。
*Fisheye および Crucible は除く
要約
Synchrony is suddenly killed by OS. In the synchrony log (atlassian-synchrony.log), you'll see something like this:
2024-10-24 20:21:11,655 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # A fatal error has been detected by the Java Runtime Environment: 2024-11-10 09:11:24,365 DEBUG [15419:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] #
2024-10-24 20:21:11,658 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # SIGSEGV (0xb) at pc=0x00007f1f864b2da4, pid=54, tid=2123
2024-10-24 20:21:11,658 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] #
2024-10-24 20:21:11,659 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # JRE version: OpenJDK Runtime Environment Temurin-11.0.23+9 (11.0.23+9) (build 11.0.23+9)
2024-10-24 20:21:11,659 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # Java VM: OpenJDK 64-Bit Server VM Temurin-11.0.23+9 (11.0.23+9, mixed mode, tiered, compressed oops, g1 gc, linux-amd64)
2024-10-24 20:21:11,659 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # Problematic frame: 2024-10-24 20:21:11,676 DEBUG [15419:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # V [libjvm.so+0x7fcda4] G1ParScanThreadState::copy_to_survivor_space(InCSetState, oopDesc*, markOopDesc*)+0x374
2024-10-24 20:21:11,676 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] #
2024-10-24 20:21:11,677 DEBUG [111:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # An error report file with more information is saved as: 2024-11-10 09:11:24,365 DEBUG [15419:StdOutHandler [/lib/jvm/jdk-11.0.23+9/bin/java]] # /var/atlassian/application-data/confluence/hs_err_pid54.log環境
Confluence Data Center
診断
The generated file, hs_err_pid54.log, will have a bit more information, but the part we're looking for is that the JVM library is failing while trying to perform G1GC:
# A fatal error has been detected by the Java Runtime Environment: #
# SIGSEGV (0xb) at pc=0x00007f1f864b2da4, pid=54, tid=2123
#
# JRE version: OpenJDK Runtime Environment Temurin-11.0.23+9 (11.0.23+9) (build 11.0.23+9)
# Java VM: OpenJDK 64-Bit Server VM Temurin-11.0.23+9 (11.0.23+9, mixed mode, tiered, compressed oops, g1 gc, linux-amd64)
# Problematic frame: # V [libjvm.so+0x7fcda4] G1ParScanThreadState::copy_to_survivor_space(InCSetState, oopDesc*, markOopDesc*)+0x374原因
Although there are reasons to take in consideration, we can note that the JVM is being killed by *G1ParScanThreadState::copy_to_survivor_space*. There is currently a bug opened for JDK to track this problem
ソリューション
The solution for this has been to change the GC algorithm for Synchrony, by adding the below parameter to the synchrony-args.properties in the confluence home, or in the shared home if it's a Cluster.
synchrony.jvm.arg.gc=-XX:+UseParallelGCこの内容はお役に立ちましたか?