Must be my day for weirdness.
I have two LCLSX agents that are hanging up the agent manager. I hae LS agents configured to run < 30 min during the day and they’re running, well, forever…
I’ve tried tell amgr quit - no joy
I’ve done the tell amgr cancel “file” ‘agent’ and get an error ‘The AMGR task either does not exist or does not accept the ‘tell’ commands’
sh ta shows the agents are running (and running and running).
Server reboot time or is there something else I can try?
TIA
Doug
Subject: Tell amgr quit - won’t
'tis time to reboot…
had you issued the amgr cancel before the quit you may have been able to recover, but not likely.
Subject: RE: Tell amgr quit - won’t
Thanks. I’ve got the reboot set for about noon.
Interestingly, the LS tasks shouldn’t be running but are and are consuming 0% resources…
Now if I can just figure out why an array won’t redim…(see earlier post)
Friday, Friday, Friday, Friday…gooooood.
Thanks again.
Doug
Subject: RE: Tell amgr quit - won’t
Since its Friday, forget about your array and just go home - looks like your having one of those days where you should have stayed in bed.
If your not at work, you won’t miss out on anything!
Hopefully once you reboot the box, the agents won’t kick off again and you might run into the same problem. Either unschedule the agents or remove the amgr from the server’s ini until you figure out what is wrong with your agents. Perhaps they are recurrsing in an endless loop.
Subject: RE: Tell amgr quit - won’t
This may be relevant, if you are using a script library:
Either way, I’m with Tom - in fact, I think we should all go home.
Subject: RE: Tell amgr quit - won’t
Kawa-bingo!
OK, so for years I fought using libraries - gotta set them up, have to call them, when you debug you’re going from code set a to the library to … I finally caved, built some libraries and got a huge chunka skin pulled off my posterior…
Thank you for the link. I feel better.
Doug
Subject: RE: Tell amgr quit - won’t
Oh I killed those agents dead…told the dbs not to run scheduled agents and disabled them for good measure.
The agents are LCLSX agents that pull data out of Oracle and stuff it into Notes. The agents have been running hourly or a couple of months and take 10 minutes or so to process.
Yesterday, I spoke to one of our Oracle gurus and we discovered that the SQL I was using probably wasn’t getting me what I needed so I changed the SQL code…this is a day I wish I had a dev server…the agents are taking hours to complete.
What I don’t understand is why they’re not aborting at the 30 minute mark defined in the server doc.
Oh well, I’m going to work on the query syntaxt and see if I can figure this one out. Worst case, I can actually kill these two dbs - they may not be able to provide the data we need anyway.
Thanks for the assist.