Post questions here relative to DataStage Enterprise/PX Edition for such areas as Parallel job design, Parallel datasets, BuildOps, Wrappers, etc.
Moderators: chulett , rschirm , roy
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Thu Feb 10, 2005 9:08 am
Hi,
I observed a weird thing today while I was running the job. I ran a job and after it loaded around 200,000 records, I have aborted the job. Later when I reset and ran the job again, the performance statstics shows 210,000 rows, 0 rows/sec. After sometime, again I aborted, rest and the ran the job again. Now, the performance statistics showed 240,000, 0 rows/sec. This clearly showed that it started reading from the point where it was previously aborted.
Any reasons for this peculiar behaviour?
Cheers,
Mohan
Sainath.Srinivasan
Participant
Posts: 3337 Joined: Mon Jan 17, 2005 4:49 am
Location: United Kingdom
Post
by Sainath.Srinivasan » Thu Feb 10, 2005 9:39 am
Sometimes when you stop / reset a job, the background engine and the datastage controller goes out of sync and hence maybe this problem. Try to cleanup resources and locks in such cases.
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Thu Feb 10, 2005 11:43 am
Hi,
When I try to clean up resources, I encounter the following message:
"Cannot find any process numbers for stages in job LOAD_PS_RO_LINE".
What does that mean?
Cheers,
Mohan
amsh76
Charter Member
Posts: 118 Joined: Wed Mar 10, 2004 10:58 pm
Post
by amsh76 » Thu Feb 10, 2005 11:51 am
Hey are you doing any aggregation in that Job ? then this kind of behaviour is quite possible..if your incoming data is not sorted..
Describe your job, please.
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Thu Feb 10, 2005 12:34 pm
Hi,
I'm not doing any aggregation. It's a plain one-to-one job (matching columns).
Cheers,
Mohan
amsh76
Charter Member
Posts: 118 Joined: Wed Mar 10, 2004 10:58 pm
Post
by amsh76 » Thu Feb 10, 2005 12:36 pm
do you have enough space ?
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Thu Feb 10, 2005 12:38 pm
Well...that's a gud question. We are lagging at space but we have enough space to accomodate data. What's the point that you are trying to make?
Cheers,
Mohan
amsh76
Charter Member
Posts: 118 Joined: Wed Mar 10, 2004 10:58 pm
Post
by amsh76 » Thu Feb 10, 2005 12:43 pm
What i am trying to say is, is it by any chance..after writing around 200000 rows..its going out of space ?
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Thu Feb 10, 2005 12:53 pm
Nopes....it didn't....
It didn't write even a record..
Cheers,
Mohan
T42
Participant
Posts: 499 Joined: Thu Nov 11, 2004 6:45 pm
Post
by T42 » Mon Feb 14, 2005 4:49 pm
Do not worry very much about the Job Monitor. It is a very inprecise tool provided as a 'visual aid' at the expense of performance.
Instead, focus on the log output for precise information.
DataStage does not 'continue' on the job level.
sathyanveshi
Participant
Posts: 66 Joined: Tue Dec 07, 2004 12:48 pm
Post
by sathyanveshi » Mon Feb 14, 2005 4:56 pm
Thanks a lot for your responses..
Cheers,
Mohan
Sainath.Srinivasan
Participant
Posts: 3337 Joined: Mon Jan 17, 2005 4:49 am
Location: United Kingdom
Post
by Sainath.Srinivasan » Tue Feb 15, 2005 4:53 am
Try resetting the job and run again. If nothing works, export the job, delete from DS and import it back again. This will remove and rebuild the job's metadata.