cassandra-commits mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Benedict (JIRA)" <>
Subject [jira] [Commented] (CASSANDRA-9998) LEAK DETECTED with snapshot/sequential repairs
Date Tue, 25 Aug 2015 14:26:46 GMT


Benedict commented on CASSANDRA-9998:

No, this message indicates that everything is now fine, but there is a programmer error that
caused a resource to not be released before it was GC'd (and was then released on GC). I guess
we should change the message that is printed to make clear to operators that there is no need
to panic, but there is a need to visit JIRA and inform us of the message. This message indicates
the possibility of otherwise available resources (including memory) being unusable for extended
periods (i.e. until GC occurs)

Could you doublecheck this is definitely the same instance of the message, and if not file
a separate bug report? The basic message is potentially universal, as it covers any number
of objects and uses across the codebase. If you could enable {{-Dcassandra.debugregcount=true}},
and check the stack traces it produces are the same (or if you could apply this fix and confirm
the errors disappear), I'd appreciate it 

> LEAK DETECTED with snapshot/sequential repairs
> ----------------------------------------------
>                 Key: CASSANDRA-9998
>                 URL:
>             Project: Cassandra
>          Issue Type: Bug
>            Reporter: Marcus Eriksson
>            Assignee: Marcus Eriksson
>             Fix For: 2.1.9, 2.2.1, 3.0 beta 1
> does not happen if I add -par to the test

This message was sent by Atlassian JIRA

View raw message