Oracle errors not generating director warnings

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

Post Reply
NigeGriff
Premium Member
Premium Member
Posts: 46
Joined: Mon Nov 24, 2003 5:46 am

Oracle errors not generating director warnings

Post by NigeGriff »

In server most oracle errors would generate a warning in the director job log.

This does not seem to happen in enterprise.

Is there a reason for this and is there a way to force oracle errors to generate a warning in enterprise.
Thanks
Nigel
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Depends what you're doing. For example, if your settings invoke sqlldr then the errors will be in the sqlldr log file (the bad file may also be populated). Once sqlldr starts, control has left DataStage, which simply waits for sqlldr to finish. Put another way, sqlldr does not emit its errors on stderr, so there's no way to capture them.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
NigeGriff
Premium Member
Premium Member
Posts: 46
Joined: Mon Nov 24, 2003 5:46 am

Post by NigeGriff »

Ray,
i'm specifically talking about using the oracle upsert feature.
Thanks
Nigel
NigeGriff
Premium Member
Premium Member
Posts: 46
Joined: Mon Nov 24, 2003 5:46 am

Post by NigeGriff »

Ray,
it is specifically around the scenario when records are rejected during the upsert. The log shows how many records have been rejected but they do not generate warnings as they would in server.

Is there a reason for this and is there a way to make director generate warnings for these rejected records in the same way that server does.

We could then set the warnings threshold to abort the job when met.
Thanks
Nigel
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

As you are finding, that's just The Way It Works. Certain errors happen silently and are only visible upon close examination of the logs. We've found you really need a routine to examine the logs after job and abort the job (or issue some kind of alert) when those conditions are found.
-craig

"You can never have too many knives" -- Logan Nine Fingers
ray.wurlod
Participant
Posts: 54607
Joined: Wed Oct 23, 2002 10:52 pm
Location: Sydney, Australia
Contact:

Post by ray.wurlod »

Does your Oracle Enterprise stage have a rejects link? If it does, I would expect DataStage to log an alert message if it is used. The reject link should also be able to deliver sqlcode automatically so you can determine why the row was rejected, at least generically.
IBM Software Services Group
Any contribution to this forum is my own opinion and does not necessarily reflect any position that IBM may hold.
chulett
Charter Member
Charter Member
Posts: 43085
Joined: Tue Nov 12, 2002 4:34 pm
Location: Denver, CO

Post by chulett »

That's the problem - it doesn't. Or to put it more properly, there are many times where data is rejected that does not make it down a reject link. Everything looks happy happy until you take a closer look at the logs. Makes much sense, it does not. :?
-craig

"You can never have too many knives" -- Logan Nine Fingers
Post Reply