[Starcluster] starcluster in loop

[Mark J. Pearrow]

Hmmm, here's a little more on this. I can launch other AMIs just fine using the same credentials and I can log in, etc. However, if I launch a starcluster AMI - either by using the starcluster python stuff or even by launching it directly from elasticfox, I see the following in the console log for the instance:

/etc/init.d/rc: 390: /etc/rcS.d/S50ec2-ssh-host-key-gen: Permission denied

and then the boot process hangs here:

* Starting periodic command scheduler crond       
[80G 
[74G[ OK ]

Since cron is init'ed at S89cron in runlevel 2, I suspect that it might be wedging on running sgeexecd - not sure why though. Maybe the key generation that failed is causing a hiccup here?

Is anyone else able to launch a starcluster AMI instance at the moment or is it just me...

mjp