[
https://issues.apache.org/jira/browse/RATIS-677?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16931174#comment-16931174
]
Lokesh Jain commented on RATIS-677:
-----------------------------------
[~szetszwo] Thanks for working on this! If we ignore the exception while
reading a segment file wouldn't that make the log segments inconsistent? If the
segment with the error or the segments after that are used later, it might lead
to unpredictable results.
> Logentry marked corrupt due to ChecksumException
> ------------------------------------------------
>
> Key: RATIS-677
> URL: https://issues.apache.org/jira/browse/RATIS-677
> Project: Ratis
> Issue Type: Bug
> Components: server
> Reporter: Sammi Chen
> Assignee: Tsz Wo Nicholas Sze
> Priority: Blocker
> Attachments: r677_20190913.patch
>
>
> Steps:
> 1. Run Teragen and generated a few GB data in a 4 datanodes cluster.
> 2. Stoped the datanodes through ./stop-ozone.sh.
> 3. Changed the ozone binaries
> 4. Start the cluster through ./start-ozone.sh.
> 5. Two datanode regisisterd to SCM. Two datanode fail to appear at SCM side.
>
> Checked these two failed node, datanode process is still running. In the
> logfile, I found a lot of following errors.
> 2019-09-12 21:06:45,255 [Datanode State Machine Thread - 0] INFO -
> Starting XceiverServerRatis ba17ad5e-714e-4d82-85d8-ff2e0737fcf9 at port 9858
> 2019-09-12 21:06:47,255 [Datanode State Machine Thread - 0] INFO -
> Attempting to start container services.
> 2019-09-12 21:06:47,255 [Datanode State Machine Thread - 0] INFO -
> Background container scanner has been disabled.
> 2019-09-12 21:06:47,255 [Datanode State Machine Thread - 0] INFO -
> Starting XceiverServerRatis ba17ad5e-714e-4d82-85d8-ff2e0737fcf9 at port 9858
> 2019-09-12 21:06:47,255 [Datanode State Machine Thread - 0] ERROR -
> Unable to communicate to SCM server at 10.120.110.183:9861 for past 2100
> seconds.
> org.apache.ratis.protocol.ChecksumException: LogEntry is corrupt. Calculated
> checksum is -134141393 but read checksum 0
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLogReader.decodeEntry(SegmentedRaftLogReader.java:299)
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLogReader.readEntry(SegmentedRaftLogReader.java:185)
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLogInputStream.nextEntry(SegmentedRaftLogInputStream.java:121)
> at
> org.apache.ratis.server.raftlog.segmented.LogSegment.readSegmentFile(LogSegment.java:94)
> at
> org.apache.ratis.server.raftlog.segmented.LogSegment.loadSegment(LogSegment.java:117)
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLogCache.loadSegment(SegmentedRaftLogCache.java:310)
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLog.loadLogSegments(SegmentedRaftLog.java:234)
> at
> org.apache.ratis.server.raftlog.segmented.SegmentedRaftLog.openImpl(SegmentedRaftLog.java:204)
> at org.apache.ratis.server.raftlog.RaftLog.open(RaftLog.java:247)
> at
> org.apache.ratis.server.impl.ServerState.initRaftLog(ServerState.java:190)
> at
> org.apache.ratis.server.impl.ServerState.<init>(ServerState.java:120)
> at
> org.apache.ratis.server.impl.RaftServerImpl.<init>(RaftServerImpl.java:110)
> at
> org.apache.ratis.server.impl.RaftServerProxy.lambda$newRaftServerImpl$2(RaftServerProxy.java:208)
> at
> java.util.concurrent.CompletableFuture$AsyncSupply.run(CompletableFuture.java:1590)
> at
> java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
> at
> java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
> at java.lang.Thread.run(Thread.java:748)
--
This message was sent by Atlassian Jira
(v8.3.2#803003)
