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

Re: [HTCondor-users] GSI_ASSIST_GRIDMAP not calling out



Hi,

I've just found a very good clue, in /var/log/messages:

Dec 21 13:24:09 hepgrid6 gsi_pep_callout[1869]: Authorizing DN /C=UK/O=eScience/OU=Liverpool/L=CSD/CN=stephen jones Dec 21 13:24:09 hepgrid6 gsi_pep_callout[1869]: argus_pep_callout: gsi_pep_callout_error: Authorization error: Can not map /C=UK/O=eScience/OU=Liverpool/L=CSD/CN=stephen jones to local identity#012gsi_pep_callout_error: PEP client error: Failed to authorize XACML request: Problem with the local SSL certificate

So it's not a condor problem. Condor is calling out properly to the library, I think.

I reckon it's something to do with set-up on the host. I should be able to debug that... I'll let you know exactly what I find.

Ste



On 20/12/18 20:46, Brian Bockelman wrote:
Ah - is it a test build of HTCondor?  There might be some hint from the cmake output as well.

Brian

On Dec 20, 2018, at 2:45 PM, sjones <sjones@xxxxxxxxxxxxxxxx> wrote:

Hi Brian,

I've tried some of these things, and a few other things, but it's still not right.

No problem; this is only a test system. I think it's a compiler option. I think GSI might be getting commented out by #ifdefs (that's one theory, anyway.)

Cheers,

Ste

On 2018-12-20 17:26, Brian Bockelman wrote:
Hi Stephen,
The lines to focus on are these:
12/20/18 17:14:03 ZKM: 2: mapret: 0 included_voms: 1 canonical_user:
GSS_ASSIST_GRIDMAP
12/20/18 17:14:03 Globus-based mapping failed; will use gsi@unmapped.
This means that it tried to invoke Globus and something failed along the lines.
Given you say it never makes it Args:
1.  Any sign it makes it to Globus libraries?
2.  Any sign it Globus successfully handed off to libgsi_pep_callout.so?
3.  Any sign the PEP callout did anything?
For (3), I believe there's a way (environment variable?  PEP config?
... I forget) to increase the logging verbosity.  That said, sometimes
I find "strace" the easiest way to determine some of these things.
Once we know where the ball gets dropped, we can target the debugging further.
Brian
On Dec 20, 2018, at 11:21 AM, Stephen Jones <sjones@xxxxxxxxxxxxxxxx> wrote:
Hi all,
I have a HTCondor-CE that contains a condor-mapfile that has GSS_ASSIST_GRIDMAP, to make a callout to ARGUS.
# cd /etc/condor-ce/
# cat condor_mapfile
GSI (.*) GSS_ASSIST_GRIDMAP
...
I've set the GSI_AUTHZ_CONF value to be right:
# echo $GSI_AUTHZ_CONF
/etc/grid-security/gsi-authz.conf
And the conf file file looks OK:
# cat /etc/grid-security/gsi-authz.conf
globus_mapping /usr/lib64/libgsi_pep_callout.so argus_pep_callout
And the library is installed:
# ls -lrt /usr/lib64/libgsi_pep_callout.so
lrwxrwxrwx 1 root root 27 Dec 20 16:22 /usr/lib64/libgsi_pep_callout.so -> libgsi_pep_callout.so.1.0.1
Yet, when I try it out, it never calls out to my ARGUS server:
12/20/18 17:14:03 ZKM: setting default map to gsi@unmapped
12/20/18 17:14:03 ZKM: name to map is '/C=UK/O=eScience/OU=Liverpool/L=CSD/CN=stephen jones'
12/20/18 17:14:03 ZKM: pre-map: current user is 'gsi'
12/20/18 17:14:03 ZKM: pre-map: current domain is 'unmapped'
12/20/18 17:14:03 ZKM: map file already loaded.
12/20/18 17:14:03 ZKM: attempting to map '/C=UK/O=eScience/OU=Liverpool/L=CSD/CN=stephen jones'
12/20/18 17:14:03 ZKM: GSI was used, and FQAN is present.
12/20/18 17:14:03 ZKM: 1: attempting to map '/C=UK/O=eScience/OU=Liverpool/L=CSD/CN=stephen jones,/dteam/Role=NULL/Capability=NULL,/dteam/NGI_UK/Role=NULL/Capability=NULL'
12/20/18 17:14:03 ZKM: 2: mapret: 0 included_voms: 1 canonical_user: GSS_ASSIST_GRIDMAP
12/20/18 17:14:03 Globus-based mapping failed; will use gsi@unmapped.
12/20/18 17:14:03 ZKM: post-map: current user is 'gsi'
12/20/18 17:14:03 ZKM: post-map: current domain is 'unmapped'
12/20/18 17:14:03 ZKM: post-map: current FQU is 'gsi@unmapped'
12/20/18 17:14:03 AUTHENTICATE: Exchanging keys with remote side.
12/20/18 17:14:03 AUTHENTICATE: Result of end of authenticate is 1.
12/20/18 17:14:03 DC_AUTHENTICATE: authentication of <138.253.178.91:13663> did not result in a valid mapped user name, which is required for this command (1112 QMGMT_WRITE_CMD), so aborting.
Does anyone know what might be stopping it?
Cheers,
Ste
--
Steve Jones                             sjones@xxxxxxxxxxxxxxxx
Grid System Administrator               office: 220
High Energy Physics Division            tel (int): 43396
Oliver Lodge Laboratory                 tel (ext): +44 (0)151 794 3396
University of Liverpool                 http://www.liv.ac.uk/physics/hep/
_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/
_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users
The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/
_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/

_______________________________________________
HTCondor-users mailing list
To unsubscribe, send a message to htcondor-users-request@xxxxxxxxxxx with a
subject: Unsubscribe
You can also unsubscribe by visiting
https://lists.cs.wisc.edu/mailman/listinfo/htcondor-users

The archives can be found at:
https://lists.cs.wisc.edu/archive/htcondor-users/


--
Steve Jones                             sjones@xxxxxxxxxxxxxxxx
Grid System Administrator               office: 220
High Energy Physics Division            tel (int): 43396
Oliver Lodge Laboratory                 tel (ext): +44 (0)151 794 3396
University of Liverpool                 http://www.liv.ac.uk/physics/hep/