DSXchange

From the Director, highlight the job, then from the menu

Job -> Clear Log

click on the auto-purge and select your options.

If you explicitly set the purge settings for a job it will not need recompilation. These settings are written directly to the job log. When you create a new job (importing or renaming an existing one), the current default project settings are "inherited" and remain unless explicitly changed.

1 - no needto recompile.
2 - The job status is stored with the log file, I doubt that there is a bug here but the description is a bit unclear to me.

I'm at 8.7 as well. My first thought was this was supposed to show the PID or JobId of the controlling process, but even when I call my jobs from a job sequence this value is 0.

I haven't checked, but perhaps it is filled when running a server job.

But we know you have MKS Toolkit - Although the conspiracy theorists might believe thata DataStage offers a portal into your system (all your data are belong to us), wiser heads will note that MKS Toolkit is installed as part of DS on Windows systems :wink: p.s. were those answers fast enough for yo...

How about "wc -l" to count the lines, assuming the file you are checking is one delimited by line feeds.

I am using this at my present site to copy datasets around (for CDC and recovery purposes), perhaps 30-60Gb per day in about 40 datasets. It is as efficient as one can get since it mainly uses the OS-level cp which is about as fool-proof as one can get.

nansekar - actually, your advice came from ShaneMuir, not Ray.

Welcome to DSXChange. When signing up the suggestion is to try the search engine to see if the same problem might have already been posted and solved. I just did an exact search for "APT_BadAlloc: Heap allocation failed" and got 63 hits. Did you try different keywords? You have run out of ...

I tried that a while ago without luck, so ended up making it a loop to as high as possible a value and then explictly exiting the loop when I hit my exit condition.

Yes.

When you schedule a job from your client director, the underlying cron or AT scheduler on the server is used, not that on the client side.

Most likely this is part of your problem. Change the autopurge for these jobs to use age rather than runs and see if the problem goes away. As I mentioned before, there were/are problems with multi-instance jobs and purge settings.

In the specifications for the column attributes in the Row Generator stage you can not only specify whether a column is null, but also what percentage nulls are to be generated. If you specify 0% then no nulls will be created.

That is where it should be. Do other client installations at your site have the file or is it missing there as well?

I believe that DBAs are there to keep databases running by answering questions like this (they have a vested interest in minimizing impact of big/lengthy operations). Do you have one available to you? The double-quotes were inserted by me to distinguish the SQL from the normal post text. The syntax ...