Re: [Condor-users] condor_g error when globus-job-run works

[Rich Bruin <rbru03@xxxxxxxxxxxxx>]

Hello again,

So, as is always the way - as soon as you ask for help you find the 
answer yourself!

I replaced the /etc/grid-security directory with one from another 
machine and everything started working correctly! I'm not sure of the 
differences unfortunately as (stupidly) I overwrote the old directory 
and all of its files so cannot compare, but I know it is not to do with 
the grid-mapfile (as there wasn't one before or after) which this error 
code often relates to.

Sorry I can't be more specific about the solution, but thanks for reading!

Rich

Rich Bruin wrote:
> Hello All,
>
> I'm trying to debug a problem we're having with a submit machine here 
> but can't find any help via google / the list archive, so hopefully 
> someone here can help!
>
> In short, I can run globus jobs using globus directly (via 
> globus-job-run etc) but running condor jobs leads to them quickly going 
> held and reporting the following error message (from the job's log file):
>
> Globus job submission failed!
>      Reason: 7 authentication failed: GSS Major Status: Authentication 
> Failed GSS Minor Status Error Chain:  init.c:499: 
> globus_gss_assist_init_sec_context_async: Error during context 
> initialization init_sec_context
>
> In more detail, I am submitting via condor_g from a Debian sarge 
> installation (2.4.27-2-386 kernel) running globus toolkit 4.0.1 and 
> condor version 6.6.10 to any of a few remote resources running various 
> versions of Linux and globus toolkits 2.4.3, 3.2.1 and 4.0.1. Simply 
> running globus-job-run type commands works fine (directed to both the 
> fork and pbs jobmanagers) but any job run via condor_g fails with the 
> above error message in the local logs.
>
> The logs on the remote machine read as follows:
>
> Notice: 5: Authenticated globus user: 
> /C=UK/O=eScience/OU=Cambridge/L=UCS/CN=richard bruin
> Notice: 0: GRID_SECURITY_HTTP_BODY_FD=6
> Notice: 5: Requested service: jobmanager-fork
> Notice: 5: Authorized as local user: rbru03
> Notice: 5: Authorized as local uid: 501
> Notice: 5:           and local gid: 501
> Notice: 0: executing /usr/local/globus/libexec/globus-job-manager
> Notice: 0: GRID_SECURITY_CONTEXT_FD=9
> Notice: 0: Child 10758 started
> Notice: 6: globus-gatekeeper pid=10838 starting at Tue Mar  7 16:25:06 2006
>
> Notice: 6: Got connection 128.232.232.27 at Tue Mar  7 16:25:06 2006
>
> Failed reading length 0
> GSS authentication failure
>      globus_gss_assist token :3: read failure: Connection closed
> Failure: GSS failed Major:01090000 Minor:00000000 Token:00000003
>
> Failure: GSS failed Major:01090000 Minor:00000000 Token:00000003
>
> Notice: 6: globus-gatekeeper pid=10839 starting at Tue Mar  7 16:25:06 2006
>
> Notice: 6: Got connection 128.232.232.27 at Tue Mar  7 16:25:06 2006
>
> Failed reading length 0
> GSS authentication failure
>      globus_gss_assist token :3: read failure: Connection closed
> Failure: GSS failed Major:01090000 Minor:00000000 Token:00000003
>
> Failure: GSS failed Major:01090000 Minor:00000000 Token:00000003
>
> Does anyone have any idea what is happening here? I have other machines 
> with near enough identical installations and they work fine, it just 
> seems to be this one client machine!
>
> Any help you could provide would be much appreciated, thanks in advance,
>
> Rich
>
> -------------------------------
> Richard Bruin
> PhD Student
> Department of Earth Sciences
> University of Cambridge
> eMinerals project www.eminerals.org
> rbru03@xxxxxxxxxxxxx
> _______________________________________________
> Condor-users mailing list
> Condor-users@xxxxxxxxxxx
> https://lists.cs.wisc.edu/mailman/listinfo/condor-users