[jboss-jira] [JBoss JIRA] Updated: (JBCACHE-805) Channel is set to AUTO_GET_STATE even if cache doesn't want initial state transfer

[Manik Surtani (JIRA)]

     [ http://jira.jboss.com/jira/browse/JBCACHE-805?page=all ]

Manik Surtani updated JBCACHE-805:
----------------------------------

          Component/s: Replication
        Fix Version/s: 2.1.0
    Affects Version/s: 1.4.0.SP1

> Channel is set to AUTO_GET_STATE even if cache doesn't want initial state transfer
> ----------------------------------------------------------------------------------
>
>                 Key: JBCACHE-805
>                 URL: http://jira.jboss.com/jira/browse/JBCACHE-805
>             Project: JBoss Cache
>          Issue Type: Bug
>      Security Level: Public(Everyone can see) 
>          Components: Replication
>    Affects Versions: 1.4.0.SP1
>            Reporter: Brian Stansberry
>         Assigned To: Manik Surtani
>             Fix For: 2.1.0
>
>
> The channel should probably only be set to AUTO_GET_STATE if getFetchStateOnStartup() returns true.  Otherwise if the node is shunned and reconnects, a full state transfer will occur.  Presumably the cache can't or doesn't want to handle that, or getFetchStateOnStartup() would not have returned true.
> One problem is what if getFetchStateOnStartup() returns false because the config uses a shared cache loader (so no persistent state transfer) and a cold cache at startup is OK (so no in-memory state transfer).  But, following reconnect the cache is no longer "cold" -- the in-memory tree now has potentially stale data.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: http://jira.jboss.com/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira