Greetings,
I have created a cluster where the sgeadmin user is not recognized as an sge user and cannot submit jobs. Here are some diagnostics illustrating that sgeadmin exists at some level in sge.
# qconf -sm // displays a list of all grid engine system managers
root
sgeadmin
# qconf -so // displays a list of all grid engine system operators
sgeadmin
# qconf -suser sgeadmin // displays the configuration of the specified user
sgeadmin is not known as user
# qconf -suserl // displays a list of all currently defined users
root
Also, the sge binary directory is not found in sgeadmin's path.
# echo $PATH
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/opt/sge6/bin/linux-x64
# su sgeadmin
$ echo $PATH
/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games
How does one get into a situation like this and how do you correct it?
Here is the configuration for my cluster (64-bit, m1.medium, size 1, spot instance).
[cluster mycluster]
KEYNAME = <myKey>
CLUSTER_SIZE = 1
CLUSTER_USER = sgeadmin
CLUSTER_SHELL = bash
NODE_IMAGE_ID = ami-999d49f0
NODE_INSTANCE_TYPE = m1.medium
AVAILABILITY_ZONE = us-east-1b
VOLUMES = <myEbsVol>
Best regards,
Steve
--
Steve Darnell
DNASTAR, Inc.
Madison, WI USA
Received on Fri Dec 21 2012 - 18:47:07 EST