[
https://issues.apache.org/jira/browse/CASSANDRA-9084?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14506005#comment-14506005
]
Benedict commented on CASSANDRA-9084:
-------------------------------------
The results of a quick hacky script to tell us which duplicate log messages
there are:
{noformat}
Validating {}
RepairJob.java:212
RepairJob.java:225
RepairJob.java:266
RepairJob.java:279
Failed to set keep-alive on Thrift socket.
TCustomNonblockingServerSocket.java:58
TCustomServerSocket.java:117
Could not configure socket.
TCustomSocket.java:75
TCustomSocket.java:125
adding secondary index {} to operation
Keyspace.java:446
Keyspace.java:491
skipping {}
KeysSearcher.java:159
CompositesSearcher.java:206
Write timeout; received {} of {} required replies
StorageProxy.java:574
StorageProxy.java:690
Exception in thread {}
CassandraDaemon.java:234
CassandraDaemon.java:235
CassandraDaemon.java:243
Skipping {}
KeysSearcher.java:143
CompositesSearcher.java:196
Loading {}
Schema.java:472
Schema.java:524
Schema.java:555
Schema.java:583
[repair #{}] {}
RepairSession.java:176
RepairSession.java:230
RepairSession.java:243
RepairSession.java:266
RepairJob.java:187
RemoteSyncTask.java:54
LocalSyncTask.java:67
LocalSyncTask.java:107
Notified {}
Gossiper.java:966
Gossiper.java:980
Read only {} (< {}) last page through, must be done
KeysSearcher.java:110
CompositesSearcher.java:167
Sleeping for {}ms to ensure {} does not change
Gossiper.java:515
Gossiper.java:591
JVM doesn't support Adler32 byte buffer access
FBUtilities.java:640
FBUtilities.java:672
DiskAccessMode is {}, indexAccessMode is {}
DatabaseDescriptor.java:315
DatabaseDescriptor.java:320
Enqueuing response to {}
RangeSliceVerbHandler.java:37
MutationVerbHandler.java:52
ReadVerbHandler.java:43
Drop {}
Schema.java:573
Schema.java:601
Updating {}
Schema.java:536
Schema.java:564
Schema.java:592
Could not initialize SIGAR library {}
SigarLibrary.java:51
SigarLibrary.java:55
Empty output skipped, filter empty tuples to suppress this warning
CassandraStorage.java:511
CqlNativeStorage.java:396
Acquiring sstable references
CollationController.java:77
CollationController.java:204
Index build of {} complete
SecondaryIndexManager.java:170
SecondaryIndex.java:222
Miscellaneous task executor still busy after one minute; proceeding with
shutdown
StorageService.java:661
StorageService.java:3827
Failed to set receive buffer size on Thrift socket.
TCustomNonblockingServerSocket.java:81
TCustomServerSocket.java:140
Snapshot-based repair is not yet supported on Windows. Reverting to parallel
repair.
StorageService.java:2793
StorageService.java:2861
Failed to set send buffer size on Thrift socket.
TCustomNonblockingServerSocket.java:69
TCustomServerSocket.java:128
{} found, but does not look like a plain file. Will not watch it for changes
GossipingPropertyFileSnitch.java:92
PropertyFileSnitch.java:79
Opening {} ({} bytes)
SSTableReader.java:384
SSTableReader.java:430
Some replicas have already promised a higher ballot than ours; aborting
StorageProxy.java:376
StorageProxy.java:410
Corrupt sstable {}; skipped
SSTableReader.java:454
SSTableReader.java:477
Could not set socket timeout.
TCustomSocket.java:139
TCustomServerSocket.java:159
Compaction executor has shut down, not submitting task
CompactionManager.java:533
CompactionManager.java:603
Cannot start multiple repair sessions over the same sstables
RepairMessageVerbHandler.java:100
CompactionManager.java:1033
Scanning index {} starting with {}
KeysSearcher.java:115
CompositesSearcher.java:172
Executor has shut down, not submitting background task
CompactionManager.java:183
CompactionManager.java:1294
Reached end of assigned scan range
KeysSearcher.java:166
CompositesSearcher.java:238
Unavailable
StorageProxy.java:582
StorageProxy.java:684
error registering MBean {}
CassandraDaemon.java:551
BlacklistedDirectories.java:54
{noformat}
I don't have a super strong opinion either way, but I do dislike adding line
numbers when they're mostly redundant. I am mostly certain under current
default scenarios it has zero performance impact, but it does leave open a
window for performance regressions if logging happens on any critical paths
without realisation, and on enabling debug/trace logging it very likely does
already, so having the defaults configured not to upset cluster behaviour too
much when enabling this is a positive. But, like I say, my position is very
weakly held.
> Do not generate line number in logs
> -----------------------------------
>
> Key: CASSANDRA-9084
> URL: https://issues.apache.org/jira/browse/CASSANDRA-9084
> Project: Cassandra
> Issue Type: Improvement
> Components: Config
> Reporter: Andrey
> Assignee: Benedict
> Priority: Minor
>
> According to logback documentation
> (http://logback.qos.ch/manual/layouts.html):
> {code}
> Generating the line number information is not particularly fast. Thus, its
> use should be avoided unless execution speed is not an issue.
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
