Home | About | Sematext search-lucene.com search-hadoop.com
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB
 Search Hadoop and all its subprojects:

Switch to Plain View
MapReduce >> mail # user >> Re: interval from Live to Dead about DataNode


+
Nicolas Liochon 2013-07-04, 08:27
Copy link to this message
-
答复: interval from Live to Dead about DataNode
Nicolas,

 

Thanks for your help. had a look at HDFS-3703.

 

So I need to turn on dfs.namenode.check.stale.datanode

and set a shorter time for dfs.namenode.stale.datanode.interval

 

Thanks,

Francis.Hu

 

发件人: Nicolas Liochon [mailto:[EMAIL PROTECTED]]
发送时间: Thursday, July 04, 2013 16:28
收件人: [EMAIL PROTECTED]
主题: Re: interval from Live to Dead about DataNode

 

Hi,
It's 10 minutes and 30s.

See the stale mode described in HDFS-3703 if you need something shorter.

Cheers,
Nicolas

 

On Thu, Jul 4, 2013 at 10:05 AM, Francis.Hu <[EMAIL PROTECTED]>
wrote:

Hi,All

 

Does anyone know when the status of datanode switch from live to dead at the
internal of namenode ?

 

the scenario:

When i stopped a datanode with command, the status of that datanode in the
web UI of namenode displays 'live' and 'In Service' for almost 5 minutes.  

I know the default value of datanode dfs.hreadbeat.interval is set as 3
seconds and dfs.namenode.decommission.interval is set 30 seconds.

 

Thanks,

Francis.Hu

 

+
Nicolas Liochon 2013-07-04, 09:14
NEW: Monitor These Apps!
elasticsearch, apache solr, apache hbase, hadoop, redis, casssandra, amazon cloudwatch, mysql, memcached, apache kafka, apache zookeeper, apache storm, ubuntu, centOS, red hat, debian, puppet labs, java, senseiDB