DSXchange

What about other sessions for the user it mentioned?

What does it show you? That session? If so, have you tried disconnecting it?

Sorry, was still in PX mode for some reason. :wink: Same advice applies, however, unless there's more to your 'not equal' than a failure of the equality check. Since you're using an OCI stage rather than a hashed file reference you can't use the NOTFOUND link variable for this check but you can fall...

So... an actual Lookup stage or something else? Would not your check for 'not equal' be true when your 'equal' check fails? Set the stage to 'Continue' and check for null output in that case.

You're confident why? Did you check the last modification date/time for both via the Manager? Is there a 'From previous run...' error message logged by the job when you Reset it?

Have you checked from the Web Console?

Is this a 'proper' multi-instance job? Meaning, does each invocation write to a separate fileset?

'No rows found' generally indicate a metadata issue, define all your fields as strings to see what's in the file. If you need help with the 'few that are generaing with invalid dates/error values' you'd need to provide quite a bit more detail before anyone could help with it.

String parameters are not 'required' and are allowed to be empty, as that is a valid value for them. The same is not true for Integer or Date fields, they must be given a valid value at run time.

I'd start here.

Caused by: com.ibm.db2.jcc.c.SqlException: [ibm][db2][jcc][t4][2012][11248] Connection authorization failure occurred. Reason: Password expired.

There's no way for anyone here to know. Something has obviously changed, the problem is going to be determining what. The shared container, perhaps?

If this topic is indeed not resolved for you, please let us know why. What questions remain? What needs to be clarified?

If all you're looking for is for someone to explain or 'define crontab' for you, there are plenty of resources out there already on the subject that Google would easily turn up.

'ls' is as much of a 'program' as cat is. The External Source stage just give you a mechanism to bring in data from an 'external source' which (as I understand it) can be anything that delivers information to standard out. And if you were using 'cat' there, to me that implies you weren't following E...

Set your warning limit appropriately when you run the job and set up a Reject link from the OE stage to capture the problem records wherever you like.

There are several stages that assume nullable output, the lookup and aggregator for example. Regardless of what you know, they will assume nullable and force you to treat it as such and handle nulls 'properly' to make them happy.