You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@hbase.apache.org by Heng Chen <he...@gmail.com> on 2016/06/14 03:35:43 UTC
Is there any way to release rpc handler when all handlers are
occupied by big request.
Currently, we found sometimes our RS handlers were occupied by some big
request. For example, when handlers read one same big block from hdfs
simultaneously, all handlers will wait, except one handler do read block
from hdfs and put it in cache, and other handlers will read the block from
cache. At this time, RS fall in stuck mode. And i have to restart the RS
to recover.
Is there any way i can release the handlers and reject the same request?
Re: Is there any way to release rpc handler when all handlers are
occupied by big request.
Posted by Heng Chen <he...@gmail.com>.
My HBase version is 0.98.17.
My key point is that should we supply one tool which could kill big request
just like mysql kill slow query?
2016-06-15 1:50 GMT+08:00 Esteban Gutierrez <es...@cloudera.com>:
> Hi Heng,
>
> That sound like some issue from older versions from HBase. Can you please
> give us a little more of details? like which version of HBase are you
> using and a stack dump from the RS?
>
> cheers,
> esteban.
>
>
> --
> Cloudera, Inc.
>
>
> On Mon, Jun 13, 2016 at 8:35 PM, Heng Chen <he...@gmail.com>
> wrote:
>
> > Currently, we found sometimes our RS handlers were occupied by some big
> > request. For example, when handlers read one same big block from hdfs
> > simultaneously, all handlers will wait, except one handler do read block
> > from hdfs and put it in cache, and other handlers will read the block
> from
> > cache. At this time, RS fall in stuck mode. And i have to restart the
> RS
> > to recover.
> >
> >
> > Is there any way i can release the handlers and reject the same request?
> >
>
Re: Is there any way to release rpc handler when all handlers are
occupied by big request.
Posted by Esteban Gutierrez <es...@cloudera.com>.
Hi Heng,
That sound like some issue from older versions from HBase. Can you please
give us a little more of details? like which version of HBase are you
using and a stack dump from the RS?
cheers,
esteban.
--
Cloudera, Inc.
On Mon, Jun 13, 2016 at 8:35 PM, Heng Chen <he...@gmail.com> wrote:
> Currently, we found sometimes our RS handlers were occupied by some big
> request. For example, when handlers read one same big block from hdfs
> simultaneously, all handlers will wait, except one handler do read block
> from hdfs and put it in cache, and other handlers will read the block from
> cache. At this time, RS fall in stuck mode. And i have to restart the RS
> to recover.
>
>
> Is there any way i can release the handlers and reject the same request?
>