The two other things to check are:
Troubles in the ~/.globus/.gass_cache on the remote host. This is the least frequent cause of this error. It could relate to NFS or AFS issues on the remote host. It is also possible that the CA that issued your Globus certificate is not trusted by your local host. Running 'grid-proxy-init -verify - debug' should detect this situation.

If you use netstat and see the port listening on the port inside the TCP port range, you can try duplicating the connection using non-job tools like netcat and/or telnet and verifying that there's nothing wrong with the hostnames/ports used.


Charles

On Jun 22, 2007, at 12:48 PM, Zheng Chen wrote:

Hi All,
There is a failure when I submit jobs to a globus enabled machine blue05
in my university.
"globus-job-run blue05.iridis.soton.ac.uk /bin/hostname".
The error is the infamous "GRAM Job submission failed because the job
manager failed to open stderr (error code 74)".

Here's what I've tried or verified:

1. Firewalls -- I've turned off my firewall and set the $GLOBUS_TCP_PORT_RANGE
[EMAIL PROTECTED] ~]$ echo $GLOBUS_TCP_PORT_RANGE
40000,41000

2. Hostname -- both hostname and globus-hostname return the correct and
qualified hostname.
[EMAIL PROTECTED] ~]$ echo $GLOBUS_HOSTNAME
zc05rvm.ecs.soton.ac.uk

3. certificates --
[EMAIL PROTECTED] ~]$ globusrun -a -r blue05.iridis.soton.ac.uk
GRAM Authentication test successful

4. I can perform GridFTP to get and put files from/to blue05
globus-url-copy file:/tmp/local_file \
    gsiftp://blue05.iridis.soton.ac.uk/tmp/zchen
is successful

5. However, there is an empty grid-mapfile in /etc/grid-security

   Can anyone tell me how to sort the problem out?
   many thanks in advance.

Zheng




Reply via email to