Harald Kirsch created KAFKA-4502:
------------------------------------
Summary: Exception during startup, append offset to swap file
Key: KAFKA-4502
URL: https://issues.apache.org/jira/browse/KAFKA-4502
Project: Kafka
Issue Type: Bug
Components: core
Affects Versions: 0.10.2.0
Environment: Windows Server
Reporter: Harald Kirsch
During startup, the Kafka server throws the exception shown below with a bit of
pre-context.
We are using the so-called SiphonRelease
(https://github.com/Microsoft/Kafka/tree/SiphonRelease,
https://issues.apache.org/jira/browse/KAFKA-1194?focusedCommentId=15702991)
which tries to circumvent problems of the logCleaner to rename and delete
segments still memory mapped by the broker.
The trouble seems to be as follows: since in the SiphonRelease the LogCleaner
still sometimes crashes, we have a monitoring script that detects this and then
restarts the Windows Service (apache procrun based) running Kafka. My hunch is
that the combination of restart-service/procrun does not allow Kafka to shut
down smoothly, since when it starts we get tons of messages like:
{noformat}
[2016-12-05 23:30:20,704] WARN Found a corrupted index file due to requirement
failed: Corrupt index found, index file
(d:\Search\kafka\fileshare-0\00000000000000084814.index) has non-zero size but
the last offset is 84814 which is no larger than the base offset 84814.}.
deleting d:\Search\kafka\fileshare-0\00000000000000084814.timeindex,
d:\Search\kafka\fileshare-0\00000000000000084814.index and rebuilding index...
(kafka.log.Log)
{noformat}
While this seems fixable by Kafka, my hunch is that a leftover .swap file then
breaks it as follows:
{noformat}
[2016-12-05 23:32:34,676] INFO Found log file
d:\Search\kafka\windream-4\00000000000000000000.log.swap from interrupted swap
operation, repairing. (kafka.log.Log)
[2016-12-05 23:32:34,957] ERROR There was an error in one of the threads during
logs loading: kafka.common.InvalidOffsetException: Attempt to append an offset
(110460) to position 182 no larger than the last offset appended (110735) to
d:\Search\kafka\windream-4\00000000000000000000.index.swap.
(kafka.log.LogManager)
[2016-12-05 23:32:34,957] FATAL Fatal error during KafkaServer startup. Prepare
to shutdown (kafka.server.KafkaServer)
kafka.common.InvalidOffsetException: Attempt to append an offset (110460) to
position 182 no larger than the last offset appended (110735) to
d:\Search\kafka\windream-4\00000000000000000000.index.swap.
at
kafka.log.OffsetIndex$$anonfun$append$1.apply$mcV$sp(OffsetIndex.scala:132)
at kafka.log.OffsetIndex$$anonfun$append$1.apply(OffsetIndex.scala:122)
at kafka.log.OffsetIndex$$anonfun$append$1.apply(OffsetIndex.scala:122)
at kafka.utils.CoreUtils$.inLock(CoreUtils.scala:234)
at kafka.log.OffsetIndex.append(OffsetIndex.scala:122)
at kafka.log.LogSegment.recover(LogSegment.scala:224)
at kafka.log.Log$$anonfun$loadSegments$5.apply(Log.scala:248)
at kafka.log.Log$$anonfun$loadSegments$5.apply(Log.scala:232)
at scala.collection.immutable.Set$Set1.foreach(Set.scala:74)
at kafka.log.Log.loadSegments(Log.scala:232)
at kafka.log.Log.<init>(Log.scala:108)
at
kafka.log.LogManager$$anonfun$loadLogs$2$$anonfun$3$$anonfun$apply$10$$anonfun$apply$1.apply$mcV$sp(LogManager.scala:151)
at kafka.utils.CoreUtils$$anon$1.run(CoreUtils.scala:58)
at
java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:511)
at java.util.concurrent.FutureTask.run(FutureTask.java:266)
at
java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1142)
at
java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:617)
at java.lang.Thread.run(Thread.java:745)
{noformat}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
