OW Services

Toni_Nanneman

Active Member
We are in the process of preparing for the Xe ugprade. Two weeks ago we
upgraded our AS/400 OS to V4R5 and this is the second week in a row we've
had to reset our OneWorld services on a Thursday morning.

Users appear to be working fine but after around 9:00 or so, new users are
getting 'Security Server' errors and are unable to connect to OneWorld. We
bump the services and then everything works fine! We IPL every Sunday so
our services are reset once a week. Additionally, we purge the SQLPKG files
on the 400 before every IPL.

Does anyone have any ideas of what could be causing this? Are there any
special problems with V4R5 and B7322? Thanks in advance for any
constructive comments!!!

Toni Nanneman
Holiday Retirement Corp.



B7322 SP12.2; NT; SQL 6.5; coexist on AS/400 with A73 cum 8
 
Although this probably seems like a weak answer and not much help. I have seen many examples where the OneWorld Kernals will fail after a long period of time and especially under heavy load. You might want to consider stopping and starting your services a bit more often.

You might also be able to extend the amount of time that your system will run by adding some additional KERNAL jobs.

I must also admit that I was not aware that SP12.2 for B7322 supported V4R5. I do know that originally there was only support for up to V4R4 but I lost track of which service pack changed that.

I wish I could be more helpful.

Good Luck.

Shane
 
I encountered a problem similar to this a while ago. One of my kernel job's
job log was filling up and the as400 would end the job. Check your qezdebug
queue for the ending kernel job and find if this is what is causing your
problem. I increased the max size parm joblog and the issue was resolved.
This is an AS400 system value and is reset when you change OS.
Good luck.

Tom Mitchell
 
Toni Nanneman,
Several years ago I worked at a company that was B7322 OneWorld,
HPUX, Oracle, NT. I believe the unstableness that you are seeing is
just a "quirt" with that release. (Just my opinion so please don't take
it wrong). :)
We saw this type of problem daily and at different times of the
day. There never seemed to be rhyme or reason to why this was
happening. We did see that this was happening due to the fact that the
number of processes running would go below 17 when this would happen.
It got to the point we wrote a cron that would bounce OneWorld if
processes went below 17. Real pain in the .... During that time I
spoke with several other CNC types who were working on this version and
they seemed to be having the same issue. Tweaking and tuning helped the
frequency, but we never got the bounce down below 2 times a week. :(

Lisa G. Stinebuck
Senior Service Delivery Technician
Logical EBOC Cincinnati
513-412-7950 x1021
[email protected]
 
Back
Top