Re: listclusters bug

From: Justin Riley <no email>
Date: Sun, 19 Dec 2010 01:23:32 -0500

On 12/7/10 11:28 AM, sg00097_at_surrey.ac.uk wrote:
> Hi, first off starcluster is amazing, so thank you.
>
> I decided to make an EBS volume using starcluster but to keep the instance that made it running afterwards, however after successfully creating the volume (although it did have a lot of fail attempt in trying to connect via ssh when doing so) I wasn't able to log into or stop the instance, and when trying the listcluster command i was presented with there error in the attachment, I have since manually closed down the instance but still get presented with the same error and now unfortunately that error is the only thing that comes up when I use the command even when I have another instance i created through starcluster running.
>
> Any help on resolving this issue would be greatly appreciated, thanks.
Sorry for the late response and that you're having issues. The problem
is the 'listclusters' command requires a 'master' node and the
createvolume command creates a cluster without a 'master' node. This is
fixed in the latest github code. Try logging in to the aws management
console (http://aws.amazon.com/console) and deleting the
_at_sc-volumecreator security group. That should fix things, let me know if
you still have issues. Until the next release, you should allow
starcluster to shutdown the instance(s) after creating the volume(s)
which will remove the group for you automatically or just remember to
remove the _at_sc-volumecreator group.

~Justin
Received on Sun Dec 19 2010 - 01:23:14 EST