Severe MRP Slowdowns / Issues

[bbandrowski (Blake Bandrowski)]

Hello all,

Our MRP has been having a lot of issues lately, our process went from taking 8-9 hours to over 15, and EpicCare hasn’t given me a lot to work with in terms of coming up with a fix. Can someone please take a look at the log and hopefully give me an idea where to start?

Thanks so much!
TVC MRP Logs - Broken.zip (9.8 MB)

[bbandrowski (Blake Bandrowski)]

EpicCare’s only real idea on how to fix this would be to run fewer regenerative runs, and fix the “uncountable” number of revisions that aren’t approved.

[Nancy_Hoyt]

HI Blake,

Is there any chance you are losing internet or network connection during the run? This seems very odd:

00:42:16 Finished Processing Part:R721906-002
00:42:36 System.Data.SqlClient.SqlException (0x80131904): A transport-level error has occurred when receiving results from the server. (provider: Session Provider, error: 19 - Physical connection is not usable)

Searching on this type of error I found this post. Perhaps it will provide some clues:

SQL Server - Transaction is in doubt error

Hello E10help, We have been having issues with our Epicor server needing to rebooted in the middle of the night (All users get an “Unsecured fault” error). I’ve actually found that it’s an issue with the SQL Service/s and not the server itself - usually if I restart the services in a correct manner then it can recover without a reboot. I’ve been looking into the logging and found that there’s usually a “Transaction is in doubt (TransactionInDoubtException)” error before seeing the issue with …

We run full regen nightly and use 5 processes and schedulers. If one were to fail the others would keep going. Maybe bump yours up to 6? We don’t fail much anymore and back when we did it was due to excessively long deletion times on transfer order suggestions and also with bad methods built with resources from multiple plants. Haven’t had those in awhile however…

HTH,
Nancy

[josecgomez]

What version of Epicor are you on?

[hkeric.wci (Haso Keric)]

Also are you still on 3rd Party Hosted, that could be an issue too. Been down this road many times, youd be surprised what caused the slowness… sometimes faulty Firewall Config, NAS, SAN, a RAID Controller being bad, VMWare misconfigurations among other things.

Jose looks like ERP11.2.300.0

Also when did this start? Im assuming you ran MRP for many months prior to this slowness without issue.

[JasonMcD (Jason McDermott)]

@Nancy_Hoyt

when we did it was due to excessively long deletion times on transfer order suggestions

Wow, this is so common! Wish I had known a year ago.

Well, I just wrote up a PPT on that, since we conquered that, I think:

Process MRP Failing

Here you go. Pick a flavor: Why is this step of MRP taking so long.pptx (690.4 KB) Why is this step of MRP taking so long.pdf (651.8 KB)

But that aside, I’d agree this issue is something else.

[bbandrowski (Blake Bandrowski)]

It is showing Kinetic 2023.1.10, we are on the public cloud!

[Nancy_Hoyt]

Wow Jason that’s an awesome detailed analysis. This is very useful for how to track down anything taking excessively long in the mrp. Thank you!

Nancy

[bbandrowski (Blake Bandrowski)]

Yea this is an awesome presentation!