You are viewing a plain text version of this content. The canonical link for it is here.

Posted to general@hadoop.apache.org by Vadim Zaliva <lo...@codeminders.com> on 2008/12/26 20:41:45 UTC

cluster machines dying

Hi!

I am experiencing a strange problem with hadoop-0.19. I have set up  
small
cluster with 4 machines. These are older machines, which have been under
heavy use for years. Once I have started to run hadoop on them they
unexplainably die after approximately 24 hours of use.

I have not even run any tasks on they yet. They are now accumulating  
some data
files, being copied in DFS. There is nothing in syslog prior to the  
crash.

Memory usage is moderate (no swap is used). CPU load is under 1. Does  
anybody seen
something like this? What steps I could take to diagnose this problem  
further?

Thanks!

Sincerely,
Vadim

P.S.

Java version: java version "1.6.0_11"

Kernel: Linux datamining1 2.6.23.1-42.fc8 #1 SMP Tue Oct 30 13:55:12  
EDT 2007 i686 i686 i386 GNU/Linux

--
"La perfection est atteinte non quand il ne reste rien a ajouter, mais
quand il ne reste rien a enlever."  (Antoine de Saint-Exupery)