DSXchange

See if this article helps. You need to get the component order correct and it can do the rest.

ps. Just use IConv - for internal format no OConv is needed since it does 'output' or 'external' format conversions.

Ah... that makes (unfortunate) sense. Good "Catch-22" to note.

Known bug with MI jobs that have 'Auto Purge' turned on. If turning that off doesn't help, see if this does.

http://www-01.ibm.com/support/docview.w ... -8&lang=en

Hmmm... if you are (or were) a SysAdmin then you should already know the answer. :? Check under /etc for the .allow and .deny files for cron and at. Add the user name to whichever is appropriate. Create them if need be. at.allow at.deny cron.allow cron.deny "cron" is used for recurring sch...

As noted a SysAdmin needs to do it, not you.

Use stage variables in a transformer to do Group Change Detection and handle the sequence number yourself - set to '1' for each new 'group' and add one while you're still in a group. Using a Sort stage set to "Don't Sort (Previously Sorted)" with "Create Key Change Column" set to...

Click on the small right-facing arrow-head next to the Property, Insert Job Parameter and then select New... if you need to create a new one for this.

You need to have a SysAdmin give the scheduling user permissions to do so. That means an entry in both cron.allow and at.allow will need to be made for each user.

Never having seen it I can't say exactly where to look, but go to the Manager / Stage Types and open up all of the branches there. If you have it installed you'll find it there. Somewhere.

OK, perhaps, but still you don't have "unlimited resources" to go along with your "unlimited connections". You will hit a limit at some point.

Sorry, but I actually have no idea what could be going on there at the moment.

I honestly don't believe that to be true, Andy. Check the generated code for a Sequence job that has "OK (Conditional)" triggers in it - it knows to check for DSJS.RUNOK for a Job Activity stage and to check for zero when using one on an Execute Command stage. At least that's what I recall...

When you said:

AGG
Grouped (A)
Sum (A)

You actually meant this, yes?

AGG
Grouped (A)
Sum (B)

And for the record, using TO_CHAR() in your source query has no time/performance penalty to speak of.

chitravallivenkat wrote:But there is no limitation on simultaneous connections on SQL server.

I find that very hard to believe.