-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi Filipe,
Ubuntu's cloudinit[1] handles mounting ephemeral drives on the
instances. StarCluster's latest 11.04 AMIs should always allocate the
maximum number of ephemeral drives available for a given instance
type. Most likely the ephemeral disks are there they're just not
mounted anywhere given that cloudinit picks the first drive available
and mounts it on /mnt. You can check this using the following command,
as root, on any StarCluster node:
$ fdisk -l
I need to investigate this but it might be possible to have
StarCluster mount these extra drives as /scratch, /scratch1,
/scratch2, etc. assuming we can reliably determine which drives are
ephemeral.
~Justin
[1]
https://help.ubuntu.com/community/CloudInit
On 08/01/2012 08:43 PM, Filipe J. Ribeiro wrote:
> Hi,
>
> I'm new to StarCluster so I might be missing something obvious.
>
> I have an instance running (m1.xlarge) that, according to amazon
> (http://aws.amazon.com/ec2/#instance) should have 1690 GB of local
> storage. I believe starcluster mounts it as /mnt/. However, /mnt/
> has only 415 GB of size.
>
> I tried an m1.small instance, which should have 160 GB of local
> store, and 'df' shows 147 GB, which is fine.
>
> Any ideas why this is happening? Is it an Amazon issue or a
> starcluster issue?
>
> Thanks for the help.
>
> Filipe
>
>
>
> _______________________________________________ StarCluster mailing
> list StarCluster_at_mit.edu
> http://mailman.mit.edu/mailman/listinfo/starcluster
>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.19 (GNU/Linux)
Comment: Using GnuPG with Mozilla -
http://enigmail.mozdev.org/
iEYEARECAAYFAlAalxgACgkQ4llAkMfDcrnpawCfU7DI8yixQsEPkdAm24qM/xxm
NlIAn2CmBHFvNoAmeEXn2EDLn11SBqjA
=L/9A
-----END PGP SIGNATURE-----
Received on Thu Aug 02 2012 - 11:05:00 EDT