cluster problem (on node)

Looking for new hardware to run WRF? Intel or AMD? Check this forum.
Post Reply
Antonix
Posts: 260
Joined: Fri Oct 16, 2009 8:53 am

cluster problem (on node)

Post by Antonix » Fri Jun 01, 2012 7:08 pm

a few days ago a strange thing happens to me. occasionally, a node (node04) hangs in my cluster. if I perform a ping, the node responds. but if I try to enter (with ssh node04) the node does not respond. you have any idea about it?? did you ever?? use ubuntu 12.04

meteoadriatic
Posts: 1601
Joined: Wed Aug 19, 2009 10:05 am

Re: cluster problem (on node)

Post by meteoadriatic » Sat Jun 02, 2012 7:40 am

Looks like ssh daemon or firewall issue. Maybe even other network problems with routing tables, with switches, with dynamically allocated IP addresses, anything...

When this happens you can try, for example, scan your node with nmap, this will reveal if everything is ok with network and wheather it listens and has open sshd port.

Post Reply