[
https://issues.apache.org/jira/browse/MAPREDUCE-1800?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12869340#action_12869340
]
Arun C Murthy commented on MAPREDUCE-1800:
------------------------------------------
FWIW the current heuristics protect reduces against a common case of a single
node (on which the map ran), and works reasonably well.
What I'm reading here is that we need better overall metrics/monitoring of the
cluster and enhancements to the masters (JobTracker/NameNode) to take advantage
of the metrics/monitoring stats. Is that reasonable?
> using map output fetch failures to blacklist nodes is problematic
> -----------------------------------------------------------------
>
> Key: MAPREDUCE-1800
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1800
> Project: Hadoop Map/Reduce
> Issue Type: Bug
> Reporter: Joydeep Sen Sarma
>
> If a mapper and a reducer cannot communicate, then either party could be at
> fault. The current hadoop protocol allows reducers to declare nodes running
> the mapper as being at fault. When sufficient number of reducers do so - then
> the map node can be blacklisted.
> In cases where networking problems cause substantial degradation in
> communication across sets of nodes - then large number of nodes can become
> blacklisted as a result of this protocol. The blacklisting is often wrong
> (reducers on the smaller side of the network partition can collectively cause
> nodes on the larger network partitioned to be blacklisted) and
> counterproductive (rerunning maps puts further load on the (already) maxed
> out network links).
> We should revisit how we can better identify nodes with genuine network
> problems (and what role, if any, map-output fetch failures have in this).
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
