hadoop-common-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Adi <adi.pan...@gmail.com>
Subject Re: Slow generation of blockReport at DataNode
Date Sun, 07 Aug 2011 15:48:43 GMT
dfs.heartbeat.interval3Determines datanode heartbeat interval in seconds.
Increasing the heartbeat interval could be an option?
OTOH delay in finding a data node is down might cause data loss.

-Adi

On Sat, Aug 6, 2011 at 5:23 PM, Joe Stein <charmalloc@allthingshadoop.com>wrote:

> Does anyone have some workarounds I can try for "Slow generation of
> blockReport at DataNode causes delay of sending heartbeat to NameNode"?  I
> see it is closed JIRA https://issues.apache.org/jira/browse/HADOOP-4584 in
> 0.21 release but looking to-do something now that will solve this without
> having to upgrade.
>
> Any folks have this that have worked around it or anyone have any ideas it
> would be greatly appreciated.
>
> I have already tried cronning  find ./ to my hdfs directories but no go.
>
> Thanks in advance.
>
> /*
> Joe Stein
> http://www.linkedin.com/in/charmalloc
> Twitter: @allthingshadoop
> */
>

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