Daily IS restarting

Our school is almost ready to go live with an integration we have developed. Briefly, this integration polls 4 network folders for text files. When it finds one, it inserts the contents into our PeopleSoft system. If an (a PS error) error occurs during those inserts, we get a document which is then e-mailed to a user that states the error.

Our problem is that each morning we have to restart our Integration Server to resume processing of the error documents.

Any idea what could cause this?

Are you running on IBM AIX? There is a known issue with the IBM JVM 1.4 causing IS lockups. What version of IS are you on?

Ah, we are running IS 6.0.1 on WinXP.

It doesn’t actually lockup in anyway. The IS continues polling and handling our flat flat files. It’s bust when an error document is published to the broker (that happens offsite), our IS will stop handling them.

Thanks,

Kelly

Kelly,

Is the trigger not picking up the queued documents from the broker?

Tell us more about the error you are receiving when publishing the error document to the broker.

There is not a specific error that occurs. The IS simply stops processing error documents that are published to the broker.

Once it is restarted, it processes them just fine. It seems that it is working fine in the evening, and when we come in the next morning, it has stopped processign them.

I’m wondering if some strange thing may be happening with our connectivity overnight.

I know there’s not much info to go on here, but thanks for any ideas you may have!

Kelly

I’d imagine if you’re losing connectivity to the broker you should have errors in the server and error log.

If you don’t have broker connectivity errors, try to force them. Create a test service which populates your publishable document with dummy data and publishes it to the broker. Schedule this so it runs every couple of minutes during the night.

I’d put a debuglog message in the beginning of the service being called by the trigger. This will determine when and if it’s running.

What is the “Interval Between Attempts” on the trigger service settings tab?