DSXchange

That's still in the queue...

Not a NULL, no - but a closer examination of the log with the dreaded 'The value of the row is' message makes it look like it might have been an empty string: <snippet> MODFN_USER = XXXXXXX LDH_CNTNT_VAL = BGN_EFF_DT = XXXXX </snippet> LDH_CNTNT_VAL is the CLOB field and it shows no value for it, ru...

You applied the recommended what? From the post:

set MAXFETCHBUFFSIZE to a value larger than its default of 8192 bytes

What did you set it to? And which uvodbc.config file did you change - the one in your Project or the one in the Engine directory?

Just curious if anyone has ever had an encounter with this Oracle error before? It turned up for the first time in one particular job the other day and I'm not really having any luck determining what might have caused this. My DBAs are a little stumped as well. We have one table with a CLOB field. M...

You mean like Oracle external tables?

'Remember me' should just be a cookie AFAIK - do you have cookies enabled?

gsbrown wrote:Thanks for the ideas.
The file presence logic works fine. I added a loop that does smaller sleep commands and checks for the presence of a flat file. If the file exists, then it loops through another sleep command, else exits the program.

I do believe he did, Master Ray.

I've seen this '141' return code from dsjob in the past, even when using the -jobstatus option. It used to crop up occasionally, and from what I recall it was only from a Sequence job that had actually Finished OK. I also think it only happened with Sequence jobs that used the Wait For File stage. I...

What version of DataStage? I believe there was a bug of this nature in earlier 7.x versions that has since been corrected...

I believe that is true but I didn't scroll all the way back up to the top to check what we were on in this thread. Still worth putting on the table for the UNIX folks out in the audience, I would think.

I believe PX jobs support USERSTATUS only if you use a BASIC Transformer. Can anyone confirm or deny?

Originally, I added a loop...repeat statement with small sleep commands and hoped pressing the job stop button would interrupt the loop between sleep commands...NOPE! It waited until the loop finished before stopping. From my experience, you have to get extremely lucky and issue the stop command in...

Right... SUP as in 'Suppress' in case anyone was wondering. But a question - with BCI code like this, don't you fall into the "works for 30 days and then fails" camp?

The only reason we've seen here so far that I recall is the write permissions issue. One question would be - what changed? Something must have on the Citrix server to cause this, perhaps they can help track that down. I'd also suggest opening a Support case to see what other issues might cause this.

Bring your timestamps in in a proper 'year first' format suitable for comparisons and you'll be fine: YYYY-MM-DD HH24:MI:SS