Re: [Starcluster] cannot start cluster with m2.4xlarge

From: Justin Riley <no email>
Date: Tue, 27 Apr 2010 11:04:20 -0400

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Damian,

> Amazon would not immediately allocate me 12 nodes so it's just 10
> m2.4xlarge's for now. I changed my configuration file accordingly
> before starting.

Did you use starcluster to launch the instances or start them manually
when the spot instance request failed? I'm actually extremely curious to
see what happens when a spot request fails as I need to know how to
handle this properly.

Currently, I'm using launch groups when starting spot instances from
starcluster which basically only fulfills the request if all instances
can be allocated.

We might want to just 'use whatever comes up' since it sounds like you
would choose to get 10 out of 12 rather than get 0 instances if not all
spot instance requests could be fulfilled.

~Justin
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.14 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAkvW/PQACgkQ4llAkMfDcrkrugCgkLtOSmh/JFfjvZzjGeU/H1LR
o3IAn26ZBD/qs151kVgXiE434TOcm0wi
=yiv7
-----END PGP SIGNATURE-----
Received on Tue Apr 27 2010 - 11:04:22 EDT