You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@kafka.apache.org by "Jason Aliyetti (JIRA)" <ji...@apache.org> on 2016/12/01 02:57:58 UTC

[jira] [Created] (KAFKA-4472) offsetRetentionMs miscalculated in GroupCoordinator

Jason Aliyetti created KAFKA-4472:
-------------------------------------

             Summary: offsetRetentionMs miscalculated in GroupCoordinator
                 Key: KAFKA-4472
                 URL: https://issues.apache.org/jira/browse/KAFKA-4472
             Project: Kafka
          Issue Type: Bug
    Affects Versions: 0.10.0.1, 0.10.1.0
            Reporter: Jason Aliyetti


The configuration "offsets.retention.minutes" is documented as being an integer.  When large values are set (i.e. Integer.MAX_VALUE), an overflow error occurs when converting from minutes to milliseconds.  For instance, setting the config value as 2147483647 results in a offsetsRetentionMs of -60000.  This means that all committed offsets are past their expiration when they are created and will be nullified on the next expiration check, which is unexpected given the type of the configuration.

The fix would be to change
"offsetsRetentionMs = config.offsetsRetentionMinutes * 60 * 1000L"
to
"offsetsRetentionMs = config.offsetsRetentionMinutes * 60L * 1000L"
in GroupCoordinator.apply().



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)