hbase-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Andrew Purtell <apurt...@apache.org>
Subject Re: RPC - Queue Time when handlers are all waiting
Date Tue, 10 Dec 2013 18:42:28 GMT
Aside from other issues being explored on this thread, "REPL IPC Server
handler N on PORT WAITING Waiting for a call (since 22 hrs, 57mins, 38sec
ago)" looks to me like ReplicationSource setting up IPC with INT_MAX for a
timeout.


On Thu, Dec 5, 2013 at 8:49 PM, Federico Gaule <fgaule@despegar.com> wrote:

> Hi,
>
> I have 2 clusters, Master (a) - Slave (b) replication.
> B doesn't have client write or reads, all handlers (100) are waiting but
> rpc.metrics.RpcQueueTime_num_ops and rpc.metrics.RpcQueueTime_avg_time
> reports
> to be rpc calls to be queued.
> There are some screenshots below to show ganglia metrics. How is this
> behaviour explained? I have looked for metrics specifications but can't
> find much information.
>
> Handlers
> http://i42.tinypic.com/242ssoz.png
>
> NumOps
> http://tinypic.com/r/of2c8k/5
>
> AvgTime
> http://tinypic.com/r/2lsvg5w/5
>
> Cheers
>



-- 
Best regards,

   - Andy

Problems worthy of attack prove their worth by hitting back. - Piet Hein
(via Tom White)

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message