You should be able to use the container's persistence directory (under ~/.globus/persisted/) to get the list of what jobs have been submitted to GRAM. It gets cleaned up if the job is destroyed, but you could probably just periodically poll it to see what's going on without having to modify the code.

If you want to modify the code, I'm sorry but I don't know where you would do that.


Charles

On Sep 18, 2008, at 2:41 PM, Wonjun Lee wrote:

Thank you for the suggestion!
I am looking for a way not in terms of user but of admin.
Even though users know where to send a job specifying the end point, but admin sometimes needs to know through log information or other information. What I am going to do is to find following output data (i.e. Address) not by using "-job-epr-output-file" option when submitting a job, by modifying the code.
--------------------------------------
<ns00:EndpointReferenceType
xmlns:ns00="http://schemas.xmlsoap.org/ws/2004/03/ addressing"><ns00:Address>https://server-A.igt.net:8443/wsrf/services/ManagedExecutableJobService </ns00:Address><ns00:ReferenceProperties><ResourceID xmlns="http://www.globus.org/namespaces/2004/10/gram/ job">2e690244-85b8-11dd-b198-0002b33f74fc</ResourceID></ ns00:ReferenceProperties><wsa:ReferenceParameters xmlns:wsa="http://schemas.xmlsoap.org/ws/2004/03/addressing"/></ ns00:EndpointReferenceType>
--------------------------------------

I am looking for a part in the code that contains Address information in
StateMachine.java

Thank you!
Wonjun

인용 Charles Bacon <[EMAIL PROTECTED]>:

If you want to keep track of what the client has requested, either
save an EPR file:
 -o, -job-epr-output-file <filename>:
       the created ManagedJob EPR will be written to the given
       file following successful submission. The file will not be
written
       if the submission fails.

Or use an RSL file when you're submitting a multijob; the job contacts
will be in the RSL file.

Unless you ask the client to keep track of where jobs went, the
information is not stored automatically.


Charles

On Sep 18, 2008, at 12:44 PM, Wonjun Lee wrote:

I know the server-A.igt.net has log information in its container.log
file. I am
sure the globus in client knows where the job is sent. In the case
of multijob,
subjobs' destination
could be revealed by subjob's endpoint. I am running an agent who
wants job's
destinations in case that job is executing a shell where another job
submissiion
to another server like server-B, is included for the purpose of job
monitoring.

I am searching the destination information first in log file and
then variable
that contains it to go back to source code.

Thank you!
Wonjun Lee

인용 Charles Bacon <[EMAIL PROTECTED]>:

I'm not sure I understand the question. Logs about the job will show
up in server-A.igt.net's container.log, not the client's.  What are
you trying to accomplish?


Charles

On Sep 17, 2008, at 11:17 PM, Wonjun Lee wrote:

Hi all,

I am submitting a job like this.

globusrun-ws -submit -Ft http://server-A.igt.net -f example.rsl

When I tried to find the destination information, "server- A.igt.net"
at
container.log after uncommenting GRAM, RFT at container-
log4j.properties file,
I was not able to find it.

Is there a way to find a destination in the log file?

Thank you very much!
Wonjun










Reply via email to