Epicor Printing Fails at Least Once a Week

Good Morning,

Can someone direct me to how Epicor Printing fails, even though the Task Agent is still “Running?” I’ve tried setting up the service to restart automatically when it stops, but when the printing queue backs up, it’s not necessarily because the agent is stopped. Ultimately, stopping and starting the agent or restarting the service is the resolution. FYI, I recycle the service daily at 5:30AM because I’ve had issues with it failing at random times, mostly overnight. Our first shift starts at 6AM and the restart helps with any failures that may have happened overnight. After adding it to a scheduled task, I’ve had a lot less restarts required, but it still does in fact fail. Because I can’t seem to get any information on why from the event viewer, my next step is to set up an alert if the Ice.SysAgentTask table grows over some specific number, letting me know the queue is backed up.

I’ve inquired about it at Insights and in User Group meetings and seems not too many people have this same issue.

Any insight would be appreciated. Ways to deal with the failures, etc.

Thanks in advance,
Eric

You read my mind on the scheduled daily restart.

All I can say is we used to have problems all the time and now it is very rare. Honestly I can’t remember the last time it happened. Probably before we upgraded from 10.1.600 to 10.2.300.

So I would guess somewhere in there they fixed the issue…

Yes, I am running 10.1.400.22. Working on upgrading servers before Epicor, but in the meantime, it’s a pain. Makes me look bad :slight_smile:

We used to have the same all the time in 10.1.400 and now we don’t either, so there is hope if you can get the upgrade through.

I did actually use a couple of BPMs that watched for signs of the queue backing up and restarted the Task Agent automatically for a while when it was a serious problem, and it kept the problem more or less under control, but they haven’t been needed for some while now for us. It was never a good solution, just there didn’t seem to be any other way of dealing with it short of having an Epicor admin available at all times ready to jump on it.

I feel like we used to have this problem on 10.1.xx but since upgrading to 10.2.xx, haven’t seen it. There’s a bunch of performance fixes in 10.2.xx in my opinion