hadoop-mapreduce-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Harsh J <ha...@cloudera.com>
Subject Re: How to configure detection of failed NodeManager's sooner?
Date Sun, 14 Feb 2016 12:49:57 GMT
You're looking for the property
"yarn.nm.liveness-monitor.expiry-interval-ms", whose default is 600000ms
(10m). This is to be set on the ResourceManager(s)' yarn-site.xml.

(X-Ref:
https://hadoop.apache.org/docs/current/hadoop-yarn/hadoop-yarn-common/yarn-default.xml#yarn.nm.liveness-monitor.expiry-interval-ms
)

On Sun, Feb 14, 2016 at 4:18 AM Caesar Samsi <mailinglist@samsi.us> wrote:

> Hello,
>
> From testing it appears detecting a failed NM node takes quite some time
> (~5 minutes)
>
> Is there a configurable setting for this? I need a more responsive cluster.
>
> Thank you, Caesar.
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: user-unsubscribe@hadoop.apache.org
> For additional commands, e-mail: user-help@hadoop.apache.org
>
>

Mime
View raw message