Erik Konrath
31st March 2003, 16:27
Hi,

Over the past three weeks we are experiencing performance problems in both "Rebuild Inventory Transactions" and "Generate MPS orders". We think that the two are related, because of the following sequence of events:

06.03. New Manufactured items added to the item table.

07.03. First time the session "Rebuild Inventory Transactions" (tdinv0250m000) runs much longer than usual (before: 12 minutes run; since 07.03.: 1h10 minutes).

21.03. The new items have been added as planned Items to the MPS.

22.03. First time MPS runs much longer than usual (before: 30 minutes; now 12 hours!!!).

We would like to identify which of the new items cause the problem. Any hints what to look at?

Thank you for any hints.

Erik

Djie-En
31st March 2003, 17:15
Hi,

Just add a datafile to the tablespace and put it on the same disk as the RollBackSegments?
Pls. what's your OS, Database?

GN

JamesV
2nd April 2003, 07:43
It may not be appropriate for the new items to be included in the planning bills. If these new parts are referenced by a variety of BOMs, and there is contention for these items if you are using multiple bshells, it could easily extend the runtime.

Also, if there was a mass load of items, were database statistics recently regenerated?

Also, if the distribution of data changed, you may have different execution plans for the queries versus when there was less data.

What is the planning horizon for these new items, perhaps your time fence is set out too far in the future.

-- Jim

Erik Konrath
2nd April 2003, 11:13
Jim,

Thank you for your response.

We run Estimated Statistics on all our Oracle tables every night.

However we have discovered that the problem might at least partly be linked to the table timps520 (Planned MPS Production Orders). We deleted that table and recreated it with just one Planned Production Order, and the runtime was normal when generating MPS for the relevant item, whereas before the deletion of the table, the runtime was very long for the same item.

Unfortunately we don't know why...

Thanks & Regards,

Erik

osoleil74
27th March 2009, 17:24
hello

i have the same problem with this session 7 minute to 12 heurs but we dont use mps we follows the solution but we dont work for me.

do you have a explanation of this

thanks
osoleil

sushil
28th March 2009, 06:46
but we dont use mps

Which Session / Process is getting Slower.!!

osoleil74
30th March 2009, 09:38
Hello


I have resolved the problem of the session Tdinv0250m000 wiht the cpu at 100 % for 12 hours. I have make a dump of the the table article tiitm001 and after i have deleted this table anc re create from the dump.

I don't no why but it's work. the session run in 12' minutes after that.

it' very strange

thanks for your help