[
https://issues.apache.org/jira/browse/ZOOKEEPER-1485?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15302771#comment-15302771
]
Martin Kuchta commented on ZOOKEEPER-1485:
------------------------------------------
I have multiple cases of this causing cores on long-running clients. See
[ZOOKEEPER-2318|https://issues.apache.org/jira/browse/ZOOKEEPER-2318] for an
example of what this looks like.
I'm not convinced that the bin-compat option above would solve the issue, at
least for the C client as it's currently implemented. The client uses a static
variable to track the XID, so expiring the session doesn't seem like it would
reset it, and the client would keep trying negative XIDs.
Simply wrapping the XID to 0 after it hits INT32_MAX seems like it would fix
the issue, since as far as I can tell, the client XID is only used for equality
comparisons to match server responses to pending completions.
> client xid overflow is not handled
> ----------------------------------
>
> Key: ZOOKEEPER-1485
> URL: https://issues.apache.org/jira/browse/ZOOKEEPER-1485
> Project: ZooKeeper
> Issue Type: Bug
> Components: c client, java client
> Affects Versions: 3.4.3, 3.3.5
> Reporter: Michi Mutsuzaki
> Assignee: Bruce Gao
>
> Both Java and C clients use signed 32-bit int as XIDs. XIDs are assumed to be
> non-negative, and zookeeper uses some negative values as special XIDs (e.g.
> -2 for ping, -4 for auth). However, neither Java nor C client ensures the
> XIDs it generates are non-negative, and the server doesn't reject negative
> XIDs.
> Pat had some suggestions on how to fix this:
> - (bin-compat) Expire the session when the client sends a negative XID.
> - (bin-incompat) In addition to expiring the session, use 64-bit int for XID
> so that overflow will practically never happen.
> --Michi
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
