Re: Question about RPC call timeout

[wang shu]

I see, thanks a lot.

2016-12-10 3:35 GMT-06:00 Guanghao Zhang :

> Yes. For idempotent operations, the same rpc call can be executed twice.
>
> 2016-12-10 7:17 GMT+08:00 wang shu :
>
> > Hi all,
> >
> > I have a very stress test on hbase by creating 1000 threads at the client
> > side, and each thread will call PUT function, and the client will throw
> out
> > some rpctimeout exception. It seems that the client will continue
> retrying
> > until the result is not NULL. Is that possible the rpc call already reach
> > the server rpc calllqueue, but still have timeout exception? If it is
> true,
> > does that mean the same rpc call will be executed twice?
> >
> > Thanks a lot for your help.
> >
> > Best,
> >
> > Shu
> >
>

Re: Question about RPC call timeout

[Guanghao Zhang]

Yes. For idempotent operations, the same rpc call can be executed twice.

2016-12-10 7:17 GMT+08:00 wang shu :

> Hi all,
>
> I have a very stress test on hbase by creating 1000 threads at the client
> side, and each thread will call PUT function, and the client will throw out
> some rpctimeout exception. It seems that the client will continue retrying
> until the result is not NULL. Is that possible the rpc call already reach
> the server rpc calllqueue, but still have timeout exception? If it is true,
> does that mean the same rpc call will be executed twice?
>
> Thanks a lot for your help.
>
> Best,
>
> Shu
>