DSXchange

Hi Eric,

"Manage Application Failover = Y" worked! Or at least the job didn't timeout.

Reconnect and disconnect are only available when the stage is used as a target.

Didn't try splitting the job, but an useful option.

Thanks,
Bob.

Hi All, Thank you for all your various contributions. I now have the same problem on a different job, except that error 3113 caused the job to abort. The job is very simple, read an oracle table and write the data in to an sql server table. No transformations, look ups, before or after sql anywhere....

The reply from the command probably needs trim(output, @FM, "A") in a User Variable stage.

Output from commands always has the @FM character appended to the end.

Hope that helps.

Bob.

Hi, Thanks for the suggestion of looking in the Sequence code in the job header. I found the routine "DSCheckPointExists" and created the routine below. However, the routine when called from the sequence always returns "START" whether running from normal or restart. As far as I c...

Hi,

Does anyone know how to detect if a sequence has been restarted?

I would like to redirect the processing in such an event.

Many thanks,
Bob.

Hi Eric, We are on DS 9.2, oracle 11g Connecting to oracle with the oracle connector stage There are no before or after sql statements We were watching the job for some time and were able to see that there were no oracle tasks running from the database consul very early on in the 30 minute window. T...

Hi Guys, I have had a couple of jobs which have finished processing, but won't finish. The jobs are quite simple, there are no look ups, joins, sorts or aggregations. The jobs write a large number of records (58m) to oracle and sql server tables and I can see that the records have been committed to ...

It might be that the comma is in the separator list (SEPLIST) or strip list (STRIPLIST) at the top of the pattern action. If it is you may need to remove it bit that might have an effect on other patterns. Also, are you sure that your new pattern is working at all? Could Orgname be populated from an...

Hi Ray,

I have been round that loop a few times but it didn't change anything, although it's the most obvious candidate.

Cheers,
Bob.

Hello,

We had the same problem and found that partitioning on the key worked, might be worth checking the partitioning once more.

You could also try running the job in a single partition and see if that fixes it and what the response times are like.

Hope that helps!

Cheers,
Bob.

Hi, I am hoping that someone can help me out here with an xmeta connection problem. When I run the script ./RunImportStart.sh I get the message: ./RunImportStart.sh Xmeta Process Meta Data XML Import Version 1.0 11/5/14 4:56:59 PM GMT INFO(500): Starting Xmeta process meta data import. Invalid usern...

Hi Ray, I tried that, but the parallel transformer doesn't recognise @FM. We also have NLS turned on and the following dsenv settings: LANG=en_US.UTF-8 LC_ALL=en_US.UTF-8 NLS_LANG=AMERICAN_AMERICA.AL32UTF8 I have also viewed the data in hex and it appears exactly as I have described so I can't see t...

I am getting some very strange behaviour when using the field function with char(254) as the separator and I wondered if anyone could shed some light on the problem. Here is the data: Col1 Debt Consolidation~~2~448 Char(254) is represented by the tilde character for clarity If I use the following ex...

Hello, I would like to understand if there is any impact on job performance (run time) between using one very large xml stage and using a number of smaller xml stages and joining the data back together with a look up. The reason I ask is because I have a large number of multivalue fields which each ...

Hi,

I have been able to establish that the xml and xsd is well formed and can be read by third party tools which can recognise the repeating elements.

So it would appear that there is something in the stage which needs to be set or unset to achieve the required result.


Many thanks,
Bob.