Heng Chen commented on HBASE-16664:

And AP is complicated so I have not found where is the logic of operation 
timeout.. I'm not sure if we have.
See CancellableRegionServerCallable.

But on submit we pass the operationTimeout. It should be writeRpcTimeout? 
Currently, we have two timeout,   one is rpcTimeout (split into read/write),  
it is used for one rpc call. 
Another one is operationTimeout which is used to control at a high level (no 
matter how many times we do retry,  the total operation time controlled by it).

I think we should respect the rules,  pass operationTimeout and do control for 
each RPC call with "rpcTimeout" and remaining of "operationTimeout",  wdyt? 

> Timeout logic in AsyncProcess is broken
> ---------------------------------------
>                 Key: HBASE-16664
>                 URL: https://issues.apache.org/jira/browse/HBASE-16664
>             Project: HBase
>          Issue Type: Bug
>            Reporter: Phil Yang
>         Attachments: testhcm.patch
> Have not checked the root cause, but I think timeout of all operations in 
> AsyncProcess is broken

This message was sent by Atlassian JIRA

Reply via email to