Mailing List Archives
Authenticated access
|
|
|
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Condor-users] file transfer problem -- 2 NICs
- Date: Fri, 8 Sep 2006 17:47:07 -0400
- From: Jim Wilgenbusch <jwilgenb@xxxxxxxxxxx>
- Subject: [Condor-users] file transfer problem -- 2 NICs
I have a rocks cluster and use condor for job submission and
distribution. My setup is a little unusually because the CM is not
on the rocks head node. This is causing some problems that I hope
someone can help me with. Here's my setup:
Rocks head node: 2 NICs. I'll use OUT to refer to the address
visible to the world and IN as the address visible inside the cluster.
The head node: Masterd and Schedd.
Compute nodes: Masterd and Startd.
CM (Collector and Negotiator) is outside the Rocks cluster.
Compute nodes are all visible to the CM and to other submit nodes via
HIGHPORT/LOWPORT range.
The rocks head node has the following config:
NETWORK_INTERFACE = OUT
BIND_ALL_INTERFACES = TRUE
With this setup I can successfully submit to any startd nodes outside
the Rocks cluster from the rocks head node and other submit nodes
(outside the rocks cluster) can successfully run on any of the Rock
cluster compute (startd) nodes.
The problem occurs only when I attempt to submit jobs from the rocks
head node with the requirement that they run on the rocks compute
nodes. In other words:
rocks_head -> outside_startd works
outside_schedd -> rocks_compute works
rocks_head -> rocks_compute DOESN'T work
After the jobs are submitted and successfully matched to one of the
rocks compute nodes I get the following error messages:
Shadow log on submit node:
9/7 18:13:02 (692.1) (11543): ERROR "Error from starter on
vm2@xxxxxxxxxxxxxxxxxxxx: Failed to transfer files" at line 597 in
file pseudo_ops.C
from user submit log file:
007 (692.002.000) 09/07 18:14:42 Shadow exception!
Error from starter on vm1@xxxxxxxxxxxxxxxxxxxx: Failed to
transfer files
Any suggestion?
Jim
--
---
**********************************************
Jim Wilgenbusch
jwilgenb@xxxxxxxxxxxx
School of Computational Science
Florida State University
150-E Dirac Science Center Library
Tallahassee, Florida 32306-4120
Voice: 850.645.0307
Fax: 850.644.0098
**********************************************