DataStage jobs are not visible

csc.datastage · Post by **csc.datastage** » Mon Jul 25, 2011 6:49 am

Hello,

While I am trying to copy one server got the NEXT_VALUE error and when I closed the DataStage designer window and opened again, suddenly the jobs are not visible. Not only for me and my colleagues desktops are also got the same problem. Only Jobs category does effected this issue and the rest are same. But found my repository from DS Aministrator by using DS_JOBS table and all the jobs are available in repository. Only we're not able to see in our desktops job palette pane.

Has anybody give an idea what was going on. Please let me know how do I approach. Do I need to re-start the server. If so, will my jobs will be totally deleted from repository? Any ideas please?

-Nag

chulett · Post by **chulett** » Mon Jul 25, 2011 6:54 am

Search the forums for DS.REINDEX for the command to run and for the very necessary caveats regarding its use.

pandeesh · Post by **pandeesh** » Mon Jul 25, 2011 6:56 am

I really don't understand what do you mean by copy one server.
However i faced the same sort of issue some months back.
I understand that the jobs are there in the repositary but it's not visible in designer client.
You can confirm the same yourself by creating the new job in the name of already existing job.
If you do like that it will say the job is already exist

So, no need to worry.

In my case i have reindexed the repositary which brings all the jobs back.
I assume you aware of REINDEX ALL command,
Make sure that no one is connected to the server while issuing reindex.

Please try this and let us know whether this resolves your issue.

csc.datastage · Post by **csc.datastage** » Mon Jul 25, 2011 7:04 am

Thanks for your prompt resond. I tried to create a new job with the existing job name and it has given the message that "...job has already exists in repository". But me and my team could not able to see jobs in our desktops. What will be the reason?

chulett · Post by **chulett** » Mon Jul 25, 2011 7:10 am

Your repository is corrupted and the fix for that has been posted.

csc.datastage · Post by **csc.datastage** » Mon Jul 25, 2011 7:16 am

Thanks for your respond. Let me know if the below steps are to be follow:

1) Close all DataStage application windows from all desktops
2) Shutdown the DataStage server.
3) Need to use DS.REINDEX command with admin priviliges from DataStage administartor COMMAND.
4) Once REINDEX done successfully start DataStage server.

Kindly confirm the step 2 and 4 are necessary to make this process.

-Nag

pandeesh · Post by **pandeesh** » Mon Jul 25, 2011 7:20 am

Those are the steps which i follow in my case.
I am not sure that 2 and 4 are really necessary.
But if you follow, there wont be any adverse impact.

chulett · Post by **chulett** » Mon Jul 25, 2011 7:21 am

You need DataStage "up" to do this, so stop and then restart it before running the fix to ensure this is happening in isolation.

csc.datastage · Post by **csc.datastage** » Mon Jul 25, 2011 7:41 am

Thanks for your prompt messages. I've exclusive access to run DS.REINDEX ALL but do not have priviliges to make server up and down. That was the reason asked whether it is mandatory to server up and down while processing this command.

-Nag

csc.datastage · Post by **csc.datastage** » Mon Jul 25, 2011 7:46 am

Sorry to ask one more clarification here.

Command "DS.REINDEX ALL" means will be effect the entired repository? or it will process only DS_JOBS category? In my desktop, "Jobs" category only could not ale to visible and rest other categories like Routines, Table Definitions, Shared Containers, etc., are fine. If I run with the option "ALL" will it be impact the rest object in the repository? Any suggestions?

-Nag

chulett · Post by **chulett** » Mon Jul 25, 2011 8:02 am

No, it's not mandatory but it is the best way to ensure exclusive access to the project in question. If you run the REINDEX command while there are any locks in place, it won't hurt anything but it also won't help... i.e. nothing will change. Hence the advice to bounce DataStage (note: not the server itself, just DataStage).

Run it with "ALL" so it can fix whatever it finds that needs fixing.

pandeesh · Post by **pandeesh** » Mon Jul 25, 2011 8:43 am

Craig,

i am curious is there any other options other than ALL for REINDEX like DS.REINDEX JOBS or DS.REINDEX ROUTINES?

Thanks

ray.wurlod · Post by **ray.wurlod** » Mon Jul 25, 2011 4:53 pm

There is no JOBS or ROUTINES options for the DS.REINDEX commands. This is primarily because there are no tables of either name.

The DS.REINDEX command is actually a script (or "paragraph") that re-indexes one of the repository tables that has indexes - all such tables have names beginning with "DS_". So it's possible to re-index DS_JOBS or DS_JOBOBJECTS or DS_ROUTINES. But how do you know which one? It's usually easier just to do them all. The ALL option simply loops through all relevant table names within the script.

If you don't have exclusive access indexes will be emptied but not re-built, so you'll be able to see even less.

chulett · Post by **chulett** » Mon Jul 25, 2011 5:06 pm

Good point on the indexes, I'd forgotten that little tidbit.