[jira] Reopened: (CASSANDRA-2115) Keep endpoint state until aVeryLongTime

["Brandon Williams (JIRA)" <ji...@apache.org>]

     [ https://issues.apache.org/jira/browse/CASSANDRA-2115?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Brandon Williams reopened CASSANDRA-2115:
-----------------------------------------


Reverted.

Though harmless, this makes it impossible to know if a node is a fatclient that has been removed, and causes repeated log messages for aVeryLongTime.

> Keep endpoint state until aVeryLongTime
> ---------------------------------------
>
>                 Key: CASSANDRA-2115
>                 URL: https://issues.apache.org/jira/browse/CASSANDRA-2115
>             Project: Cassandra
>          Issue Type: Improvement
>          Components: Core
>    Affects Versions: 0.7.1
>            Reporter: Brandon Williams
>            Assignee: Brandon Williams
>            Priority: Minor
>             Fix For: 0.7.2
>
>         Attachments: 2115.txt
>
>
> In CASSANDRA-2072 we changed the gossiper so it holds onto endpoint state until QUARANTINE_DELAY has elapsed.  However, if node X is leaving and node Y goes down and stays down longer than QUARANTINE_DELAY after X has left, Y will return thinking X is still a member of the cluster.  Instead, let's hold onto the endpoint state even longer, until aVeryLongTime which is currently set to 3 days. 

-- 
This message is automatically generated by JIRA.
-
For more information on JIRA, see: http://www.atlassian.com/software/jira