How to change timeout behavior

[Caesar Samsi <ma...@samsi.us>]

Hi there,

The default timeout waiting to fetch a file is too long (it’s likely about 1 minute).

The default before the cluster releases a node is down and takes the node out is even longer (about 5 minutes).

Where are all the timeout configuration locations that can be set (and documentation hopefully)?

A specific example is simply retrieving a result text file when a node is down.

Using hdfs dfs -cat /somedir/resultfile.txt takes at least 1 minute before fetching its replica and continues.

My application needs to be pretty nimble / responsive, so this is a real blocker.

Help!

Many Thanks, Caesar.
---------------------------------------------------------------------
To unsubscribe, e-mail: user-unsubscribe@hadoop.apache.org
For additional commands, e-mail: user-help@hadoop.apache.org