see also:
   http://oscar.sourceforge.net/notes.php#item0

check to see if pbs is running, example from my test cluster of a single
head node w/ 3 compute nodes.

   [root@node0]# /sbin/service pbs_server status
   pbs_server (pid 881) is running...
   [root@node0]# cexec -c "/sbin/service pbs_mom status"
   pbs_mom (pid 521) is running...
   pbs_mom (pid 521) is running...
   pbs_mom (pid 520) is running...

make sure that the value that 'hostname' returns is the first thing listed
in "/etc/hosts" for your head node, example node0 for 10.0.0.254 below:
   [root@node0]# hostname
   node0
   [root@node0]# more /etc/hosts
   # Do not remove the following line, or various programs
   # that require network functionality will fail.
   127.0.0.1       node0.external  localhost.localdomain   localhost
   10.0.0.254      node0   lui_oscar       nfs_oscar       pbs_oscar
   10.0.0.1        node1
   10.0.0.2        node2
   10.0.0.3        node3

--tjn

 _________________________________________________________________________
  Thomas Naughton                                      [EMAIL PROTECTED]
  Research Associate                                   (865) 574-3932


On Mon, 29 Oct 2001, Ravi Srinivasan wrote:

> I was able to install the OS on the lone test client. However while
> executing test_cluster the following error was displayed.
>
> ./test_cluster 1 1
> Running a simple PBS shell job...
> qsub: Bad UID for job execution
> - simple test taking too long - there is
>    probably a problem with PBS, exiting
>
> Ravi
> ----- Original Message -----
> From: "Thomas Naughton" <[EMAIL PROTECTED]>
> To: "Ravi Srinivasan" <[EMAIL PROTECTED]>
> Cc: <[EMAIL PROTECTED]>
> Sent: Sunday, October 28, 2001 8:46 PM
> Subject: Re: [Oscar-users] Client Installation
>
>
> > ravi,
> >
> > i'm pretty sure this is a ramdisk issue.  i had the same issue the other
> > day.  my quick fix was to use all SCSI or all IDE.
> >
> > there is a method for doing this but i'm not certain about it.  i'll bet
> > if you look in "/tftpboot/10.0.0.1/etc/lilo.conf" you'll see that it's
> > referring to sdaX (SCSI) instead of hdaX (IDE).
> >
> > i'll inquire further tomorrow but hopefully someone else on the list can
> > answer this question in shorter order.
> >
> > --tjn
> >
> >  _________________________________________________________________________
> >   Thomas Naughton                                      [EMAIL PROTECTED]
> >   Research Associate                                   (865) 574-3932
> >
> >
> > On Sun, 28 Oct 2001, Ravi Srinivasan wrote:
> >
> > > The server is all SCSI and the nodes have IDE drives.
> > >
> > > R
> > > ----- Original Message -----
> > > From: "Thomas Naughton" <[EMAIL PROTECTED]>
> > > To: "Ravi Srinivasan" <[EMAIL PROTECTED]>
> > > Sent: Sunday, October 28, 2001 8:27 PM
> > > Subject: Re: [Oscar-users] Client Installation
> > >
> > >
> > > > how different/similar are your head_node/compute_nodes?   Does one
> have
> > > > SCSI and the other not, etc?
> > > >
> > > > --tjn
> > > >
> > > >
> _________________________________________________________________________
> > > >   Thomas Naughton
> [EMAIL PROTECTED]
> > > >   Research Associate                                   (865) 574-3932
> > > >
> > > >
> > > >
> > >
> > >
> >
> >
> >
> >
> > _______________________________________________
> > Oscar-users mailing list
> > [EMAIL PROTECTED]
> > https://lists.sourceforge.net/lists/listinfo/oscar-users
> >
>
>
> _______________________________________________
> Oscar-users mailing list
> [EMAIL PROTECTED]
> https://lists.sourceforge.net/lists/listinfo/oscar-users
>



_______________________________________________
Oscar-users mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/oscar-users

Reply via email to