[
https://issues.apache.org/jira/browse/CASSANDRA-12004?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15341868#comment-15341868
]
Paulo Motta commented on CASSANDRA-12004:
-----------------------------------------
Thanks for the patch! Even though this is quite a trivial change, this has been
like this for a while and people may be relying on this format for
parsing/processing (you never know), so I think it's better to restrict this
change to trunk, since it's quite easy to fix on older versions by just
updating logback configuration if necessary.
We should probably take this chance to also update the {{STDOUT}} format to
match {{SYSTEMLOG}} format with {{ISO8601}} date and also print the thread
name. If you agree, could you make an update trunk patch with these suggestions
and also add a {{NEWS.txt}} entry so people are aware of the format change when
upgrading?
> Inconsistent timezone in logs
> -----------------------------
>
> Key: CASSANDRA-12004
> URL: https://issues.apache.org/jira/browse/CASSANDRA-12004
> Project: Cassandra
> Issue Type: Bug
> Reporter: Jérôme Mainaud
> Priority: Trivial
> Fix For: 2.1.x
>
> Attachments: patch.txt
>
>
> An error in provided logback.xml lead to inconsistent timestamp usage in logs.
> In log files, local time zone is used.
> On the console, UTC time zone is used (and millisconds are missing.)
> Example, the same log line (Local time zone: CEST) :
> in system.log
> {code}
> INFO [main] 2016-06-14 14:01:51,638 StorageService.java:2081 - Node
> localhost/127.0.0.1 state jump to NORMAL}}
> {code}
> in console
> {code}
> INFO 12:01:51 Node localhost/127.0.0.1 state jump to NORMAL
> {code}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
