[Condor-users] condor_submit -remote schedd not working...

[Erik Aronesty <erik@xxxxxxx>]

When I run a condor command from our head node (ea-bio-f0), it works
fine (we use USE_NFS), and we've run a lot of jobs with no troubles.

I set up FS_REMOTE authentication, and it seems to be OK.

However, if I attempt to use

condor_submit -remote ea-bio-f0

from another node on the system, I get this result:

----CUT---
-- Submitter: ea-bio-f0.expressionanalysis.local : <XXX:57691> :
ea-bio-f0.expressionanalysis.local
---
307125.000:  Request is held.

Hold reason: Error from slot1@xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx:
Failed to open '/var/lib/condor/spool/cluster107125.proc0.subproc0/.condor_out.9407'
as standard output: No such file or directory (errno 2)
----CUT---

So, I found this file on the head node (ea-bio-f0)

/var/lib/condor/spool/cluster107125.proc0.subproc0/.condor_out.9407

But not on the executing node (ea-bio-c12)

Why would it create it in the wrong place?

Shouldn't condor create that spool file on the executing node... not
on the schedd?