[ https://issues.apache.org/jira/browse/CASSANDRA-13503?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16000615#comment-16000615 ]
James Ravn commented on CASSANDRA-13503: ---------------------------------------- {noformat} Cassandra 2.1.17 java -version: java version "1.8.0_112" Java(TM) SE Runtime Environment (build 1.8.0_112-b15) Java HotSpot(TM) 64-Bit Server VM (build 25.112-b15, mixed mode) uname -a (Ubuntu 16.04.2): Linux ip-10-50-194-251 4.8.0-51-generic #54~16.04.1-Ubuntu SMP Wed Apr 26 16:00:28 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux lstopo: Machine (16GB total) NUMANode L#0 (P#0 7998MB) + Package L#0 + L3 L#0 (45MB) + L2 L#0 (256KB) + L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0 PU L#0 (P#0) PU L#1 (P#1) NUMANode L#1 (P#1 8047MB) + Package L#1 + L3 L#1 (45MB) + L2 L#1 (256KB) + L1d L#1 (32KB) + L1i L#1 (32KB) + Core L#1 PU L#2 (P#2) PU L#3 (P#3) {noformat} > Segfault during compaction > -------------------------- > > Key: CASSANDRA-13503 > URL: https://issues.apache.org/jira/browse/CASSANDRA-13503 > Project: Cassandra > Issue Type: Bug > Components: Compaction > Environment: {noformat} > Cassandra 2.1.17 > java -version: > java version "1.8.0_112" > Java(TM) SE Runtime Environment (build 1.8.0_112-b15) > Java HotSpot(TM) 64-Bit Server VM (build 25.112-b15, mixed mode) > uname -a (Ubuntu 16.04.2): > Linux ip-10-50-194-251 4.8.0-51-generic #54~16.04.1-Ubuntu SMP Wed Apr 26 > 16:00:28 UTC 2017 x86_64 x86_64 x86_64 GNU/Linux > lstopo: > Machine (16GB total) > NUMANode L#0 (P#0 7998MB) + Package L#0 + L3 L#0 (45MB) + L2 L#0 (256KB) + > L1d L#0 (32KB) + L1i L#0 (32KB) + Core L#0 > PU L#0 (P#0) > PU L#1 (P#1) > NUMANode L#1 (P#1 8047MB) + Package L#1 + L3 L#1 (45MB) + L2 L#1 (256KB) + > L1d L#1 (32KB) + L1i L#1 (32KB) + Core L#1 > PU L#2 (P#2) > PU L#3 (P#3) > {noformat} > Reporter: James Ravn > Attachments: hs_err_pid29774.log > > > One of our cassandra nodes segfaulted. I've attached the hs_err.log. It looks > like it was loading sstables from disk: > {noformat} > Current thread (0x00007f3df4e7b930): JavaThread "CompactionExecutor:2" > daemon [_thread_in_Java, id=30023, > stack(0x00007f3dce922000,0x00007f3dce963000)] > siginfo: si_signo: 11 (SIGSEGV), si_code: 1 (SEGV_MAPERR), si_addr: > 0xffffffffbd49a8b6 > Stack: [0x00007f3dce922000,0x00007f3dce963000], sp=0x00007f3dce961430, free > space=253k > Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native > code) > J 11168 C2 > org.apache.cassandra.utils.MergeIterator$OneToOne.computeNext()Ljava/lang/Object; > (48 bytes) @ 0x00007f3de6097df4 [0x00007f3de6097cc0+0x134] > J 2042 C2 com.google.common.collect.AbstractIterator.hasNext()Z (65 bytes) @ > 0x00007f3de560fbc0 [0x00007f3de560fb00+0xc0] > J 14662 C2 > org.apache.cassandra.db.ColumnIndex$Builder.buildForCompaction(Ljava/util/Iterator;)Lorg/apache/cassandra/db/ColumnIndex; > (36 bytes) @ 0x00007f3de7143a08 [0x00007f3de7143440+0x5c8] > J 13756 C2 > org.apache.cassandra.io.sstable.SSTableRewriter.append(Lorg/apache/cassandra/db/compaction/AbstractCompactedRow;)Lorg/apache/cassandra/db/RowIndexEntry; > (119 bytes) @ 0x00007f3de6a98b88 [0x00007f3de6a985c0+0x5c8] > J 14627 C2 org.apache.cassandra.db.compaction.CompactionTask.runMayThrow()V > (1622 bytes) @ 0x00007f3de710eed4 [0x00007f3de710daa0+0x1434] > J 18294 C1 > org.apache.cassandra.db.compaction.CompactionTask.executeInternal(Lorg/apache/cassandra/db/compaction/CompactionManager$CompactionExecutorStatsCollector;)I > (19 bytes) @ 0x00007f3de7981aa4 [0x00007f3de79818c0+0x1e4] > J 18503 C1 > org.apache.cassandra.db.compaction.AbstractCompactionTask.execute(Lorg/apache/cassandra/db/compaction/CompactionManager$CompactionExecutorStatsCollector;)I > (39 bytes) @ 0x00007f3de7a4cf2c [0x00007f3de7a4ce20+0x10c] > J 17908 C2 > org.apache.cassandra.db.compaction.CompactionManager$BackgroundCompactionCandidate.run()V > (182 bytes) @ 0x00007f3de789e0a4 [0x00007f3de789aca0+0x3404] > {noformat} > No errors in kernel logs, and no other noticeable issues on the node. > We're using {{offheap_objects}}, could that be related? Cassandra logs show > the usual memtable flushing. -- This message was sent by Atlassian JIRA (v6.3.15#6346) --------------------------------------------------------------------- To unsubscribe, e-mail: commits-unsubscr...@cassandra.apache.org For additional commands, e-mail: commits-h...@cassandra.apache.org