Scheduling through Datastage

[shalini11]

Hi All ,

when I am trying to schedule the job from datastage director, I am getting the below error.

"Cannot find record JOBPARAMINSTS in executable job file RT_CONFIG62 "

I tried to recompile the job but still the same problem persists. This problem is only in one project. In other projects on the same server, I am able to schedule the jobs.

How do I know that the RT_CONFIG file is corrupt or not and how can I rectify it.

One more question, is every project has unique RT_CONFIG file and is there only one RT_CONFIG file per project?

Can any one suggest any possible solution?

Thnx in advance

[miwinter]

Every job has an RT_CONFIG

[ray.wurlod]

You can test for a corrupted RT_CONFIG62 using UVFIXFILE RT_CONFIG62 command.

The JOBPARAMINSTS record in RT_CONFIG62 will not be created until that particuar job (job number 62 in DS_JOBS) is actually scheduled. Is the job showing in schedule view in Director?

What happens when you execute this command from your Administrator client command window?

Code: Select all

LIST.ITEM RT_CONFIG62 JOBPARAMINSTS

[ray.wurlod]

You can test for a corrupted RT_CONFIG62 using UVFIXFILE RT_CONFIG62 command.

The JOBPARAMINSTS record in RT_CONFIG62 will not be created until that particuar job (job number 62 in DS_JOBS) is actually scheduled. Is the job showing in schedule view in Director?

What happens when you execute this command from your Administrator client command window?

Code: Select all

LIST.ITEM RT_CONFIG62 JOBPARAMINSTS

[dspxguy]

Hi All ,



Can any one suggest any possible solution?

unschedule the job, recompile it and again add to scheduler.

[shalini11]

I tried with the below command:

$DSHOME/bin/uvfixfile RT_CONFIG62

The output of the command was:

"No filename specified."

Actually RT_CONFIG62 is a directory in which 3 files are there:

DATA.30, OVER.30 and one hidden file .Type30.

When I am using the same command for these files then also I am getting the same message.

Can you pls suggest what to do?

Thanks

[ray.wurlod]

You can test for a corrupted RT_CONFIG62 using UVFIXFILE RT_CONFIG62 command, executed from the Administrator client Command window.

[shalini11]

Thanks for your help.I am sorry but I am not able to understand your point. How can I run the command on the client when file is on the server....

Can you pls explain how to do this?

[chulett]

The Administrator client connects to the server. Why not just give it a try?

[shalini11]

I tried to execute the command in Administrator client and got the below output.

Beginning TRACE of RT_CONFIG62.
TRACE of RT_CONFIG62 completed.

Scanning overflow buffers.
Scan complete.

2 group (s) processed.
2 group buffer (s) processed.
6 record (s) processed.
Number of data bytes = 2508

But still I am not able to schedule.

[Jeppe]

Take a copy of the job, compile the new job, and try scheduling that.
It usually helps if all else fails

[shalini11]

Thanks a lot,

Now I am able to schedule the job. That UVFIXFILE command has worked.

[ray.wurlod]

No it hasn't.

All you have done with UVFIXFILE is a trace of RT_CONFIG62 to prove that it is not internally corrupted.

That, in turn, meant that when you recompiled the job, correct details could be written into RT_CONFIG62 and when you scheduled the job the JOBPARAMINSTS record could also be written to RT_CONFIG62 to hold the scheduling and job parameter value details.

Please don't for a moment imagine that UVFIXFILE is a silver bullet that will cure all repository ills. It definitely is not.

[shalini11]

I havent recompiled the job as this error was not coming only for a particular job. It was occuring when i was trying to schedule any job. If UVFIXFILE has not worked then I dont know how it got rectified but now everything is fine.

Thanks once again

[chulett]

What happened in the two hours between when the UVFIXFILE did not work (diagnostic only) and then suddenly things were working again?