Datastage Jobs Running Very Slow

amitendra_amit · Post by **amitendra_amit** » Thu Jul 23, 2009 10:22 am

All of sudden all the datajobs in our production environment has started running slow.
We have made no changes to any of the jobs in last 2 months.No virus scan is running on the server.Disks,memory,CPU usage all seems ok to us.
Any idea what could be the reaso behind is highly appreciated.

Cheers

ArndW · Post by **ArndW** » Thu Jul 23, 2009 10:24 am

Database speed? Network load?

Sainath.Srinivasan · Post by **Sainath.Srinivasan** » Thu Jul 23, 2009 10:25 am

Errors ? Logs ?

amitendra_amit · Post by **amitendra_amit** » Thu Jul 23, 2009 10:32 am

Database speed is fine,all the tables are analyzed and stats are gathered on weekly basis.Network load is also same as usual.There are no error or log which suggests anything fishy.
The problem is with all the jobs.

major · Post by **major** » Thu Jul 23, 2009 10:34 am

In your job sequencer is there any significant time difference in triggering the next job activity ?
If so please clear your &PH& folder except .uvnlsmap when no job is running

Sainath.Srinivasan · Post by **Sainath.Srinivasan** » Thu Jul 23, 2009 1:51 pm

amitendra_amit wrote:All of sudden all the datajobs in our production environment has started running slow.

Sudden ??

amitendra_amit · Post by **amitendra_amit** » Thu Jul 23, 2009 2:32 pm

I cleared the &PH& folder and ran some jobs but I dont see any improvement.There is no delay in triggering the jobs from sequencer.Atleast as far as I can see in director.

Yeah it started suddenly couple of days ago, the jobs which were taking 3 mins previously are now taking 15 mins.The problem is not specific to few jobs but to all the jobs

.

IBM Analytics Champion 2009 - 2020 · Post by **asorrell** » Thu Jul 23, 2009 4:34 pm

It is very doubtful that increased job runs across the board are caused by problems internal to DataStage. However, you can stop and re-start the DataStage engine just to be sure.

It could be numerous things including:
1) Contention on the Database / Database Server
2) Network traffic
3) resource constraints on any of the systems (memory / CPU)

If you are on a multi-tier system, first pin it down to a system if you can by pointing to or running jobs from a different system. Once you've got it pinned down to a system then it is just old fashioned admin / dba detective work.

djanapar · Post by **djanapar** » Fri Jan 07, 2011 11:45 am

amitendra_amit wrote:I cleared the &PH& folder and ran some jobs but I dont see any improvement.There is no delay in triggering the jobs from sequencer.Atleast as far as I can see in director.

Yeah it started suddenly couple of days ago, the jobs which were taking 3 mins previously are now taking 15 mins.The problem is not specific to few jobs but to all the jobs .

Have you got this issue resolved .If so what did you do .I am seeing same issue in my environment

ray.wurlod · Post by **ray.wurlod** » Fri Jan 07, 2011 4:16 pm

Welcome aboard.

So what's suddenly changed in your environment? Hint: "nothing" is clearly not the correct answer.

PaulVL · Post by **PaulVL** » Sat Jan 08, 2011 9:55 pm

Are you writing your logs to the metadata repository or the universe database?

djanapar · Post by **djanapar** » Mon Jan 10, 2011 11:46 am

We are having Universe database .This is a Windows server .I am checking with Intel to see if they have applied any windows patches but as far as datstage no deploymnets or no change to any jobs .

After setting up the environment variables sequencer is taking 20 minutes to start the job .Once the job starts it finishes in seconds .It is just hangs at that specific point 20 minutes for every job . I have to try clearing the &PH& file .

PaulVL · Post by **PaulVL** » Mon Jan 10, 2011 3:45 pm

Note: If it's windows, don't select the Unix platform on the message thread.

has your degrees of parallelism changed?

DSXchange

Datastage Jobs Running Very Slow

Datastage Jobs Running Very Slow

Re: Datastage Jobs Running Very Slow

Re: Datastage Jobs Running Very Slow