MRP taking excessive amount of time to run

Yahoo Archive
1
Hi Chris,

We are currently on 904.506B and are getting to go live next month. We have been running in training mode and then acceptance test mode.

Back in April when we were on (I think) 9.04.506 our test database MRP which had been completing overnight in 26 minutes suddenly went up to a couple of hours and then sometimes "hanging" and never finishing at all. We have had an ongoing support call with Epicor over this since then. I am pleased to admit that we have been much more stable over the past two months, however, with much less mrp failure or significant processing time. When we go live, I'm keeeping my fingers crossed!!

In summary, we changed our processors and schedulers many times, always a diff number depending on who we spoke to. Right now, with our stable runs we are at 2 and 2. When we have troubles, Looking at our server logs, we get a "deadlocking" error and it always seems to be associated with the deletion of unfirm jobs. We never have a long run or a failed run after a stop and restart of the appserver. When we were at the height of our problems, we started toying with doing this every day prior to mrp... hopefully we don't have to do that, but..

Our support guy at Epicor who was decent is Matt Taber, Technical Support Analyst
Vantage/Vista 8.0x: 1-800-426-8468 Epicor 9: 1-888-Epicor9
Tel: 952-417-5367
He seems knowledgeable, but we still have never gottn closure on it.


I know that a couple of months ago, I read one of the service patch release details where it gives the info on the fixes and "enhancements" and it said something about MRP getting altered so that unfirm jobs would not be renumbered. I cannot recall which version of the program it was for, but I suspect somewhere in the 9.05.. documentation. I thought hmmm I wonder if there is something they know about this unfirm job deletion causing deadlock errors.

Good Luck,
Hope this helps,
Nancy
2
Hi All

Our MRP has started taking a considerable amount of time to run - much longer
than usual.

We tend to run it a couple of times during the working day so this increase in
time adds a significant extra time per week.

What could cause the MRP to take longer to run all of a sudden?

Thanks,


CHRIS




[Non-text portions of this message have been removed]
3
We have found that we can not run MRP during the normal working day. If users are in a Job, Sales Order, Part, etc., that MRP needs, it will just hang up. We run at night, when activity is much lower, and use 15 processors for MRP and 10 more for scheduling. This insures that if some of the processors "give up the ghost", at least there are enough left to finish the job. We run a complete Regen nightly, and use the PO Suggestions or the Production Planning Process during the day, if necessary.

Rick
--- In vantage@yahoogroups.com, Chris Thompson <chriselectrix@...> wrote:
>
> Hi All
>
> Our MRP has started taking a considerable amount of time to run - much longer
> than usual.
>
> We tend to run it a couple of times during the working day so this increase in
> time adds a significant extra time per week.
>
> What could cause the MRP to take longer to run all of a sudden?
>
> Thanks,
>
>
> CHRIS
>
>
>
>
> [Non-text portions of this message have been removed]
>
4
Hi Chris. Consult the log after running - see if there's a problem - a circular reference?? I haven't seen a circular reference for well over a decade though - as I'm pretty sure business processes remove them these days.

We don't use MRP anymore, after going "lean" - implemented in an oxymoronic way I'm afraid :-( - but back in the day we'd always consult the log. I used to run a little collection of batch files and macros on the log to get rid of all the "normal" activity, filtering down to only the anomalies were left.

What version of Vantage/Epicor are you on?

->Rick - that seems like an extreme setup. Are you running a huge database? Perhaps I don't remember rightly, but I'm sure our MRP runs were easily able to be run within the 30min at lunch time when needed - and we have an undercooked server by current recommendations, V8.03.405A, and we always did a full regen (we never trusted anything less).

Kerry.

--- In vantage@yahoogroups.com, "rick.stannard" <rick.stannard@...> wrote:
>
>
> We have found that we can not run MRP during the normal working day. If users are in a Job, Sales Order, Part, etc., that MRP needs, it will just hang up. We run at night, when activity is much lower, and use 15 processors for MRP and 10 more for scheduling. This insures that if some of the processors "give up the ghost", at least there are enough left to finish the job. We run a complete Regen nightly, and use the PO Suggestions or the Production Planning Process during the day, if necessary.
>
> Rick
> --- In vantage@yahoogroups.com, Chris Thompson <chriselectrix@> wrote:
> >
> > Hi All
> >
> > Our MRP has started taking a considerable amount of time to run - much longer
> > than usual.
> >
> > We tend to run it a couple of times during the working day so this increase in
> > time adds a significant extra time per week.
> >
> > What could cause the MRP to take longer to run all of a sudden?
> >
> > Thanks,
> >
> >
> > CHRIS
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>
5
We are on 9.04.505c




________________________________
From: Kerry_Muntz <kerry_muntz@...>
To: vantage@yahoogroups.com
Sent: Thu, 2 December, 2010 20:23:45
Subject: [Vantage] Re: MRP taking excessive amount of time to run

Â
Hi Chris. Consult the log after running - see if there's a problem - a circular
reference?? I haven't seen a circular reference for well over a decade though -
as I'm pretty sure business processes remove them these days.


We don't use MRP anymore, after going "lean" - implemented in an oxymoronic way
I'm afraid :-( - but back in the day we'd always consult the log. I used to run
a little collection of batch files and macros on the log to get rid of all the
"normal" activity, filtering down to only the anomalies were left.

What version of Vantage/Epicor are you on?

->Rick - that seems like an extreme setup. Are you running a huge database?
Perhaps I don't remember rightly, but I'm sure our MRP runs were easily able to
be run within the 30min at lunch time when needed - and we have an undercooked
server by current recommendations, V8.03.405A, and we always did a full regen
(we never trusted anything less).

Kerry.

--- In vantage@yahoogroups.com, "rick.stannard" <rick.stannard@...> wrote:
>
>
> We have found that we can not run MRP during the normal working day. If users
>are in a Job, Sales Order, Part, etc., that MRP needs, it will just hang up. We
>run at night, when activity is much lower, and use 15 processors for MRP and 10
>more for scheduling. This insures that if some of the processors "give up the
>ghost", at least there are enough left to finish the job. We run a complete
>Regen nightly, and use the PO Suggestions or the Production Planning Process
>during the day, if necessary.
>
> Rick
> --- In vantage@yahoogroups.com, Chris Thompson <chriselectrix@> wrote:
> >
> > Hi All
> >
> > Our MRP has started taking a considerable amount of time to run - much longer
>
> > than usual.
> >
> > We tend to run it a couple of times during the working day so this increase
>in
>
> > time adds a significant extra time per week.
> >
> > What could cause the MRP to take longer to run all of a sudden?
> >
> > Thanks,
> >
> >
> > CHRIS
> >
> >
> >
> >
> > [Non-text portions of this message have been removed]
> >
>







[Non-text portions of this message have been removed]