+1
My team has been developing and testing against the HDFS-2832 branch for the
past month. It has proven to be quite stable.
Eric
-Original Message-
From: Arpit Agarwal [mailto:aagar...@hortonworks.com]
Sent: Monday, December 02, 2013 7:07 PM
To: hdfs-dev@hadoop.apache.org; common-..
This blog post should clear it up:
http://hortonworks.com/blog/an-introduction-to-hdfs-federation/
-Original Message-
From: ch huang [mailto:justlo...@gmail.com]
Sent: Wednesday, November 20, 2013 12:26 AM
To: hdfs-dev@hadoop.apache.org
Subject: is block pool on NN or DN?
hi,all:
I should have mentioned that the IOException is from the DataNode (if not
obvious from the stacktrace :)). The DFSClient continues along happily by
fetching the block from a different replica (as one would expect).
-Original Message-
From: Sirianni, Eric [mailto:eric.siria
scribing the problem that you saw.
>>We
>> probably can handle this better, and a jira would be a good place to
>> discuss the trade-offs. A few possibilities:
>>
>> Log a warning if mapred.submit.replication < dfs.replication.
>> Skip resetting replication if mapred.submit.
In debugging some replication issues in our HDFS environment, I noticed that
the MapReduce framework uses the following algorithm for setting the
replication on submitted job files:
1. Create the file with *default* DFS replication factor (i.e.
'dfs.replication')
2. Subsequently alter
