Run "grid-cert-diagnostics" on both machines. I think you'll find the
first machine has ef2a2c8a.0 in its trusted certs directory, and the
second machine doesn't.
Charles
On Feb 23, 2009, at 2:59 PM, Samir Khanal wrote:
Hi Charles, list
I had successfully installed GT according to the Quick start guide
Ans was able to do the gridftp work my first try.
Today after 3-4 days, i am trying to do that again.
and i get
[skha...@comet ~]$ myproxy-logon -s protos
Error authenticating: GSS Major Status: Authentication Failed
GSS Minor Status Error Chain:
globus_gss_assist: Error during context initialization
OpenSSL Error: s3_clnt.c:894: in library: SSL routines, function
SSL3_GET_SERVER_CERTIFICATE: certificate verify failed
globus_gsi_callback_module: Could not verify credential
globus_gsi_callback_module: Can't get the local trusted CA
certificate: Untrusted self-signed certificate in chain with hash
ef2a2c8a
So i again tried the 2nd part of machine setup steps at
http://www.globus.org/toolkit/docs/latest-stable/admin/quickstart/#q-second
and i am not able to log in.
Protos is the main node (elephant) and comet is the (cognito)
I am just stuck!
Samir
________________________________________
From: [email protected] [[email protected]
] On Behalf Of Samir Khanal [[email protected]]
Sent: Friday, February 20, 2009 10:38 AM
To: [email protected]
Subject: [gt-user] globus-job-get-output gives nothing
Hi all
I am sending this email again as I did not get any response..
initially the job was not executing at all, i fixed the queue and
the job now completes in the Toruqe/Maui scheduler.
The probelm now is that I cannot see the output of the execution by
using
globus-job-submit protos.cs.bgsu.edu/jobmanager-pbs /bin/hostname
i used
globus-job-get-output https://protos.cs.bgsu.edu:33541/24096/1234893868/
But that just waits and the
server_logs says
02/17/2009 13:04:39;0080;PBS_Server;Req;req_reject;Reject reply
code=15001(Unknown Job Id), aux=0, type=StatusJob, from [email protected]
02/17/2009 13:04:39;0080;PBS_Server;Req;req_reject;Reject reply
code=15001(Unknown Job Id), aux=0, type=LocateJob, from [email protected]
as the directory Globus had created inside $HOME/.globus/jobs/
protos.cs.bgsu.edu/ are all gone.
Any pointers here?
Your help will be greatly appreciated
Thanks
Samir
________________________________________
From: Charles Bacon [[email protected]]
Sent: Tuesday, February 17, 2009 11:44 AM
To: Samir Khanal
Cc: [email protected]
Subject: Re: [gt-user] My first globus-job-submit/jobmanager-pbs
script but error
The fastest thing to do is edit the pbs.pm file so that it saves the
script it is trying to qsub. Then qsub it by hand and go into a
debugging loop there to figure out why it's not matching, then fix
pbs.pm so it works for you.
Alternatively, there should be some PBS logs about why the job isn't
getting matched with a worker, so you could go check those out to see
if they show you why; perhaps it's missing a required queue attribute
or the like, or asking for the wrong number of processors, etc.
Charles
On Feb 17, 2009, at 10:42 AM, Samir Khanal wrote:
Any pointers on this yet?
-----Original Message-----
From: [email protected] [mailto:[email protected]
] On Behalf Of Samir Khanal
Sent: Monday, February 16, 2009 1:43 PM
To: [email protected]
Subject: [gt-user] My first globus-job-submit/jobmanager-pbs script
but error
Hi All
I was trying to get globus going (did it finally!)
I followed all the steps in
http://www.globus.org/toolkit/docs/4.2/4.2.1/admin/quickstart/
the Quickstart guide (Wonderfully written!)
I am able to run simple commands like echo etc but problem is with
job submission with PBS on the same machine.
$globus-job-run protos.cs.bgsu.edu /bin/hostname
these kind of scripts work.
but when i do
[...@protos ~]$ globus-job-submit protos.cs.bgsu.edu/jobmanager-pbs /
bin/hostname
https://protos.cs.bgsu.edu:58840/24878/1234808956/
[...@protos ~]$ globus-job-status
https://protos.cs.bgsu.edu:58840/24878/1234808956/
PENDING
[...@protos ~]$ qstat
Job id Name User Time Use
S Queue
------------------------- ---------------- --------------- --------
- -----
15.protos STDIN skhanal 0
Q default
The Job is always in pending mode "Q"
$pbsnodes also shows that none of the nodes are used.
Am i missing something?
The Globus is build with jobmanager-pbs and jobmanager-condor.
Your help will be greatly appreciated!
Thanks
Samir
