[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Condor-users] jobs stuck in queue

Em terça-feira 30 agosto 2011, às 16:31:41, você escreveu:
> On 08/30/2011 03:20 PM, Fabricio Cannini wrote:
> > Em quarta-feira 24 agosto 2011, às 14:48:49, David Brodbeck escreveu:
> >> On Tue, Aug 23, 2011 at 9:21 PM, Koller, Garrett
> > 
> > <kollerg14@xxxxxxxxxxxx>wrote:
> >>> Because of this, perhaps the '/var/spool/condor/' directory or one of
> >>> its subdirectories needs to be owned by root:root.  I have mine owned
> >>> by condor:condor, though, so I don't know why this is a problem.  Try
> >>> chowning it to 'root:root' and see if that helps.
> >> 
> >> If this is an NFS mount, and it's being exported without the
> >> "no_root_squash" flag, root on the client will be mapped to the user
> >> "nobody" on the server end.  This may be what you're running into.  The
> >> idea is to protect the server from rogue clients, who could otherwise
> >> pretend to have root privileges.
> > 
> > Hi everybody.
> > 
> > Any other idea on this?
> > 
> > I'm seriously pulling my hair on this.
> When you pasted the file permissions I didn't see a "." indicating
> possible context information, and I hate to suggest it because it is so
> rarely the case these days, but you might check your
> /var/log/audit/audit.log.

I guess that the closest to '/var/log/audit/audit.log' in a Debian system 
would be '/var/log/auth.log' or '/var/log/user.log' . Found nothing 

> I didn't see mention of the operating system you are using in the thread.

Sorry for this lapse. I'm using Debian amd64 6.0 on all three machines.
Packages came from here :

> It is also strange that your log indicates /var/spool/condor and your
> posted config does not suggest you set SPOOL yourself.

No, i didn't  ( manually ), it was configured to '/var/spool/condor' when i  

Is there any other info i can provide to help diagnose the root cause of it ?