I'm having the same problem :-/ Maps are going fine while reduce phase stalls on 9-16%, and then resumes after a loong while (30-40 minutes).
I'm using hadoop 0.16.0 (r618351) and wordcount hadoop-example... next week I'll try with a newer hadoop version (perhaps trunk) to see if I can reproduce this issue :-S On Sat, Jun 21, 2008 at 8:28 AM, Arnie Horta <[EMAIL PROTECTED]> wrote: > Hello, > > > I am having a problem...when I configure more than one node on my hadoop > cluster, the reduce jobs all stall. Ther logs look EXACTLY like the logs > described by Amit Kumar Singh. in his post last month. I have checked that > the disk is fine (I formatted the HDFS and deleted all the data on the data > nodes to make sure) and this is not a firewall issue or a problem with > etc/hosts. It works fine with one node, but adding a second node starts the > stalling. > > > I have tried using both .16.2 and .16.4, to no avail..... > > > HELP! >
