This just started yesterday but when we login to Epicor we have no problems, we close the program and get a 7175 and then a 7224 error, then epicor seems to shut it’s self down and when I look in progress explorer tool, the mfgsys database has stopped. If I re-start the database I am able to login to epicor again with no problems, but as soon as I try to exit out of epicor, I get the errors again and the db shuts down again. This occurs on the appserver itself and also client machines. I have went thru and disabled all BPMs active in Live and still get the errors. I have tried restarting the OE service and also all of the appservers several times. Server restart also did not fix the issue. All of the drives on the server have plenty of free space.
Any ideas? I have searched for these errors on here and it seems answers were kind of all over the place and no one had a clear resolution.
Any help is appreciated. Thanks
If this has just started happening, start with the basics, like what has changed between the time the server was working ok and now.
It may be that you have some misconfigured network equipment/dns that might be causing the issue.
Appservers show they are still running in explorer tool after the errors and the program crashes. Once I re-start the database only, I can log back into Epicor. I never touch the appservers.
Here is the broker log from the main mfgsys appserver
IO Exception received on client connection…Are you sure you don’t have some other device on your network with the same IP, or a flakey secondary DNS server, or a faulty cable.
Just to add to this. It seems that our pilot and test environments do not crash and throw the errors. So I don’t think it’s a network issue with the server. Something corrupt in the Live database perhaps?
There are a few answerbooks on EpicWeb for these error codes
Seem to point towards network issues.
11352MPS - bandwidth issues - some (new) device saturating your network?
10872MPS - network issues - TBD cause of disconnects
12497MPS - etc…
So… I might start looking at my network/devices
But… the fact that your database actually stopped has me wondering if something else is going on. Can you back the db up and do some maintenance? truncate, rebuild indexes? (I assumed this is Progress)
Do you have people actively making connections to the pilot and test environments?
Not currently. I was the only one trying those environments. I have advised users to login to test to see what the system does.
What other basic stuff have you already done
e.g. restarted appservers and then checked status details?
or maybe checked your Windows Server for an unusual number of stale user sessions?
or many large/old files in you work directories? e.g. .DBI
and checked system monitor for long running tasks? I remember having some weird problems with custom dashboard/baq hanging - where a user ran it wide open, without any parameters.
No open or hung PIDs. I also looked in system monitor and did not see any hung reports, BAQs, or processes.
Possible to backup live and restore it to either test/pilot?
And able reproduce the issue on the restored version?
Ha, yes I did just do this about 30mins ago and the issue did move over to Pilot. So that told me something is corrupt with Live since Pilot was ok using an older copy of our Live database prior to my restore. I did restore our Live DB from a backup before this happened and the errors on exit have went away and all clients have been able to login to Epicor and then exit with no errors so far. Not sure what caused the corruption as we didn’t do anything to the database.
In the Test system…
Any active labor trans? I’ve seen them cause some weird behavior in V8.
And I still might truncate the bi file and/or rebuilding indexes.
Glad to hear you are back and running. Apologies for sending you on a bit of a goose chase.