You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@zookeeper.apache.org by "Dheeraj Agrawal (JIRA)" <ji...@apache.org> on 2011/07/18 23:20:57 UTC
[jira] [Resolved] (ZOOKEEPER-1126) state of zk_handle should NOT be
initialized to 0 (CLOSED) in zokeeper_init. It should have a not
initialized state.
[ https://issues.apache.org/jira/browse/ZOOKEEPER-1126?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]
Dheeraj Agrawal resolved ZOOKEEPER-1126.
----------------------------------------
Resolution: Duplicate
Assignee: Dheeraj Agrawal
> state of zk_handle should NOT be initialized to 0 (CLOSED) in zokeeper_init. It should have a not initialized state.
> --------------------------------------------------------------------------------------------------------------------
>
> Key: ZOOKEEPER-1126
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1126
> Project: ZooKeeper
> Issue Type: Bug
> Components: c client
> Affects Versions: 3.3.3
> Reporter: Dheeraj Agrawal
> Assignee: Dheeraj Agrawal
>
> In zoo_add_auth, we have following check.
> 2954 // [ZOOKEEPER-800] zoo_add_auth should return ZINVALIDSTATE if
> 2955 // the connection is closed.
> 2956 if (zoo_state(zh) == 0) {
> 2957 return ZINVALIDSTATE;
> when we do zookeeper_init, the state is initialized to 0 and above we check if state = 0 then throw exception.
> There is a race condition where the doIo thread is slow and has not changed the state to CONNECTING, then you end up returning back ZKINVALIDSTATE from zoo_add_auth.
> The problem is we use 0 for CLOSED state and UNINITIALIZED state. in case of uninitialized case it should let it go through.
> Is this intentional? In java we have the uninitialized state = null.
> If not we can initialize it to some other magic number.
--
This message is automatically generated by JIRA.
For more information on JIRA, see: http://www.atlassian.com/software/jira