ldapsearch -x -H ldap://yourCE.yourDomain:2135 -b "mds-vo-name=local,o=grid"

"yourCE.yourDomain" should be replaced with the hostname of your computing element.

Here is an example of query results for the Computing Element "lxshare0247.cern.ch".

ldapsearch -x -H ldap://yourSE.yourDomain:2135 -b "mds-vo-name=local,o=grid"

"yourSE.yourDomain" should be replaced with the hostname of your storage element.

Here is an example of query results for the Storage Element "lxshare0206.cern.ch".

ldapsearch -x -H ldap://lxshare0246.cern.ch:2170 -b "mds-vo-name=local,o=grid"

Currently, the Resource Broker you will use is located at CERN. Here is an example of query results.

$ globus-job-run yourMachine.yourDomain /bin/hostname
yourMachine

here, "yourMachine.yourDomain" should be the hostname of your Storage or Computing Element (if you installed it)

$ globus-url-copy file://`pwd`/testfile gsiftp://yourMachine.yourDomain/tmp/testfile

$ globus-job-run yourMachine.yourDomain /bin/ls -al /tmp/testfile
-rw-rw-r--    1 lcg      lcg             0 Feb 26 15:01 /tmp/testfile

here, "yourMachine.yourDomain" should be the hostname of your Storage or Computing Element (if you installed it)

To submit a simple job, hostname.jdl, to your grid system, create the jdl script of the job as follows:

cat > hostname.jdl << EOF
Executable = "/bin/hostname";
Stdoutput = "message.txt";
StdError = "stderror";
OutputSandbox = {"message.txt","stderror"};
EOF

Before submitting your job, you can use dg-job-list-match to check if there are resources which satisfy your jobs as follows:

$ dg-job-list-match hostname.jdl

Connecting to host lxshare0246.cern.ch, port 7771

***************************************************************************
                         COMPUTING ELEMENT IDs LIST
 The following CE(s) matching your job requirements have been found:

- lxshare0247.cern.ch:2119/jobmanager-pbs-workq

***************************************************************************

Then, you can submit job to your grid:

$ dg-job-submit -o jobid -r yourCE.yourdomain/jobmanager-pbs-yourQueue hostname.jdl

Connecting to host lxshare0246.cern.ch, port 7771
Logging to host lxshare0246.cern.ch, port 15830

================================= dg-job-submit Success ==================================
 The job has been successfully submitted to the Resource Broker.
 Use dg-job-status command to check job current status. Your job identifier (dg_jobId) is:
 https://lxshare0246.cern.ch:7846/137.138.145.56/142247266396403?lxshare0246.cern.ch:7771
 The dg_jobId has been saved in the following file:
 /home/dqing/jobid
===========================================================================================

remember to replace "yourCE.yourDomain/jobmanager-pbs-yourQueue" with your Computing Element and related queue. You can use dg-job-status to chech the status of your job.

$ dg-job-status -i jobid


Retrieving Information from LB server https://lxshare0246.cern.ch:7846
Please wait: this operation could take some seconds.


*************************************************************
BOOKKEEPING INFORMATION:

Printing status info for the Job : https://lxshare0246.cern.ch:7846/137.138.145.56/142247266396403?lxshare0246.cern.ch:7771

Some bookkeeping information has not reached the LB server yet.
Missing information should come from GlobusJobmanager

---

dg_JobId                =    https://lxshare0246.cern.ch:7846/137.138.145.56/142247266396403?lxshare0246.cern.ch:7771
Status                  =    OutputReady
Last Update Time (UTC)  =    Wed Feb 26 14:19:40 2003
Job Destination         =    yourCE.yourDomain:2119/jobmanager-pbs-yourQueue
Status Reason           =    terminated
Job Owner               =    /C=TW/O=AS/OU=IP/CN=Di Qing/Email=dqing@phys.sinica.edu.tw
Status Enter Time (UTC) =    Wed Feb 26 14:19:40 2003
*************************************************************

When the status of your job is "OutputReady", you can get the result:

$ dg-job-get-output -i jobid


*****************************************************************************************************
                                JOB GET OUTPUT OUTCOME

 Output sandbox files for the job:
 - https://lxshare0246.cern.ch:7846/137.138.145.56/142247266396403?lxshare0246.cern.ch:7771
 have been successfully retrieved and stored in the directory:
 /tmp/142247266396403

*****************************************************************************************************

And

$ cat /tmp/142247266396403/message.txt
The_machine_name_where_job_is_executed

cat > replica.jdl << EOF

Executable = "/bin/sh";
Stdoutput = "stdout";
StdError = "stderror";
InputSandbox = {"testrc.sh"};
OutputSandbox = {"stdout","stderror"};
Arguments = "testrc.sh file672.txt yourSE.yourDomain/flatfiles/lcg";
EOF

here, "yourSE.yourDomain" should be the full hostname of your Storage Element, "/flatfiles/lcg" should be the VO storage area on your Storage Element, and also remember to replace filename "file672.txt" with your filename. Use an editor to create the following shell script testrc.sh:

-------- cut here --------
#!/bin/bash
#
#
echo "This is a very simple test for data replica manager tools within a job." >> $1
echo "This job will create a file, $1, and copy it to your SE $2." >> $1
echo "I execute the Replica Manager command"
edg-replica-manager-copyAndRegisterFile -l $1 -s `hostname``pwd`/$1 -d  $2/$1 -c /opt/edg/etc/lcg/rc.conf
exit 0
-------- cut here ----------

Now try to submit this job to your grid system with

$ dg-job-submit -o jobid -r yourCE.Domain/jobmanager-pbs-yourQueue replica.jdl

$ dg-job-status -i jobid

$ dg-job-get-output -i jobid

To check the actual presence of the file on the Storage Element:

$ globus-job-run yourSE.Domain /bin/ls -la /flatfiles/lcg

To query the LCG Replica Catalog, currently located at CERN, use:

ldapsearch -x -H ldap://lxshare0205.cern.ch:9011 -b \
 "lc=LCGCollection,rc=LCGRC,dc=lxshare0205,dc=cern,dc=ch"