[sheepdog] [PATCH 1/3] zookeeper: retry zk_create_seq_node on retryable error

[Kai Zhang]

On May 30, 2013, at 11:09 PM, Kai Zhang <kyle at zelin.io> wrote:

> 
> On May 30, 2013, at 11:04 PM, Liu Yuan <namei.unix at gmail.com> wrote:
> 
>> On 05/30/2013 10:58 PM, MORITA Kazutaka wrote:
>>> How do you think we handle ZINVALIDSTATE?
>>> 
>>> I've also experienced ZINVALIDSTATE and ZSESSIONEXPIRED, but, IIUC,
>>> those errors are caused by a timeout.  Am I wrong?
>>> 
>>> Either way, I think the error should be handled in the different patch
>>> unless it is a retryable one.
>> 
>> I have no idea what ZINVALIDSTATE is nor if it is retryable. But with
>> zookeeper v1.4.5, zk only return ZINVALIDSTATE for timeout as documented.
>> 
> 
> This is not retryable. Because zhandle is not valid anymore.
> In this case, we have to use zookeeper_init() to recreate a zhandle and also recreate EPHEMERAL node.
> 

Here is a document about zookeeper error handling.

http://wiki.apache.org/hadoop/ZooKeeper/ErrorHandling

Thanks,
Kyle
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.wpkg.org/pipermail/sheepdog/attachments/20130530/963340ed/attachment-0004.html>