[StarCluster] master node not in qstat

[Justin Riley]

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

Hi Adam,

> Setup a cluster today (0.93.2) and suddenly noticed that the
> 'master' node was not being reported in a "qstat -f" command and
> was not accepting run jobs from the queue . . . i.e., with 12 nodes
> x 8 cpus each (96), when 96 jobs are submitted, only 88 run (nodes
> 1-11) while 8 remain in the queue waiting.
> 
> I tried restarting the cluster using the 'sge' plugin to manually
> ensure that master_is_exec_host was set to 'True'. But the result
> was the same: 88 running - 8 waiting.

Thanks for testing and reporting. I can confirm this bug and have
created an issue on github[1]. I should have a hotfix release 0.93.3
out tonight specifically to fix this.

> But this brings up a future request. I would like to be able to run
> a cluster of 8-core servers, but have the MASTER as a non_exec node
> BUT with a different configuration (simple 2-cores, m1.large) just
> to handle file and job monitoring tasks independent of the cluster
> activity. Anyway, I know you've put more work than I can imagine
> into configuring and maintaining this package. I'm deeply
> appreciative of your skills and dedication. So I don't want to seem
> ungrateful by requesting a feature that is more of a luxury than
> anything else. Just file it aside.

You can already customize the master node's instance type and image id
by setting the following in your cluster config:

[cluster smallcluster]
MASTER_INSTANCE_TYPE=m1.large
MASTER_IMAGE_ID = ami-#######

In general MASTER_INSTANCE_TYPE and MASTER_IMAGE_ID default to
NODE_INSTANCE_TYPE and NODE_IMAGE_ID respectively if not specified.

You can also specify these at command line:

$ starcluster start -I m1.large -m ami-####### mycluster

HTH,

~Justin

[1] http://web.mit.edu/star/cluster/issues/89
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.17 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iEYEARECAAYFAk9ajGsACgkQ4llAkMfDcrnCVACglYclA7OARtZ4kmAM+Q6x1fgd
niAAnijfZS3dppfsLqHQ4lKbe5llgoYK
=2SeS
-----END PGP SIGNATURE-----