StarCluster - Mailing List Archive

Updating starcluster when spot instances go down

From: Ryan Golhar <no email>
Date: Mon, 29 Jul 2013 11:35:49 -0400

Hi all - So I've had on a few occasions nodes on my cluster get terminated
due to spikes in spot prices. When this happens, if I try to readd nodes
to my cluster, SGE (and StarCluster) and jobs are left in
the queue, nodes can't be added or perform nfs mounts, etc, etc.

Right now, the only option I have is to re-create the cluster. I think
this is overkill. Is there a way, or would people find it beneficial, if
there was a command to sync starcluster with what nodes are actually
running and update config files on the master node?

Received on Mon Jul 29 2013 - 11:35:51 EDT
This archive was generated by hypermail 2.3.0.


Sort all by: