For the past several days our nightly run of ‘Process MRP’ has failed. Historically we were running it in (full) regenerate mode, and that step was taking roughly about 7 hours. Combine that with a few other tasks in our process set and the entire run was lasting 9-10 hours. So we decided to start running MRP in ‘Recycle MRP Jobs’ mode during the week (overall execution duration dropped to 4 hours) and only run full recycle on the weekends.
A few weeks back we updated to 10.2.300.40. Since then we’ve had several executions where the ‘Process MRP’ step stops at 1-2 hours. System monitor offers no detail. The server event viewer essentially just says ‘PROGRAM CANCELLED’. So I’m having difficulty identifying the cause for the interruption.
Seems like a lot for Full Regen… I’ve dealt for months with MRP and Optimization… we got it down from 12hrs to 3hrs… few adjustments to the CPU and Hardware Remember Spartan Motors - their BOMs for a fire truck are huge.
Back to the Question:
You will have to enable MRP Verbose logging.
Basically:
MRP Logs
Task Agent Logs (ServerLog must be enabled in Admin Console)
Event Viewer on TA Windows Server running MRP
SQL Log (SQL ERROR Log would show Deadlock if it has any)
Last night logging was turned off inadvertently (I was attempting removing a step from our normal process set that has worked in the past) and we only keep the previous night’s logs. That said, normally when the interrupt is due to a part, the system monitor indicates which part. The past few nights have produced the following signs:
22:04:06 Execution Timeout Expired. The timeout period elapsed prior to completion of the operation or the server is not responding.
The statement has been terminated.
22:04:06 The wait operation timed out
22:04:06 Execution Timeout Expired. The timeout period elapsed prior to completion of the operation or the server is not responding.
The statement has been terminated.
22:04:06 at System.Data.SqlClient.SqlConnection.OnError(SqlException exception, Boolean breakConnection, Action1 wrapCloseInAction) at System.Data.SqlClient.TdsParser.ThrowExceptionAndWarning(TdsParserStateObject stateObj, Boolean callerHasConnectionLock, Boolean asyncClose) at System.Data.SqlClient.TdsParser.TryRun(RunBehavior runBehavior, SqlCommand cmdHandler, SqlDataReader dataStream, BulkCopySimpleResultSet bulkCopyHandler, TdsParserStateObject stateObj, Boolean& dataReady) at System.Data.SqlClient.SqlCommand.FinishExecuteReader(SqlDataReader ds, RunBehavior runBehavior, String resetOptionsString, Boolean isInternal, Boolean forDescribeParameterEncryption, Boolean shouldCacheForAlwaysEncrypted) at System.Data.SqlClient.SqlCommand.RunExecuteReaderTds(CommandBehavior cmdBehavior, RunBehavior runBehavior, Boolean returnStream, Boolean async, Int32 timeout, Task& task, Boolean asyncWrite, Boolean inRetry, SqlDataReader ds, Boolean describeParameterEncryptionRequest) at System.Data.SqlClient.SqlCommand.RunExecuteReader(CommandBehavior cmdBehavior, RunBehavior runBehavior, Boolean returnStream, String method, TaskCompletionSource1 completion, Int32 timeout, Task& task, Boolean& usedCache, Boolean asyncWrite, Boolean inRetry)
at System.Data.SqlClient.SqlCommand.InternalExecuteNonQuery(TaskCompletionSource1 completion, String methodName, Boolean sendToPipe, Int32 timeout, Boolean& usedCache, Boolean asyncWrite, Boolean inRetry) at System.Data.SqlClient.SqlCommand.ExecuteNonQuery() at Erp.Internal.MR.MrpExp.deleteUnfirmJob(String vPlantList) in C:\_Releases\ERP\UD10.2.300.40\Source\Server\Internal\MR\MrpExp\MrpExp.cs:line 1988 at Erp.Internal.MR.MrpExp.main_block(List1 ttMrpProcRows, List`1 ttMrpQueueRows) in C:_Releases\ERP\UD10.2.300.40\Source\Server\Internal\MR\MrpExp\MrpExp.cs:line 2644
at Erp.Internal.MR.MrpExp.RunProcess(Int64 instanceTaskNum, String outputFileName) in C:_Releases\ERP\UD10.2.300.40\Source\Server\Internal\MR\MrpExp\MrpExp.cs:line 922
22:04:06 The wait operation timed out
22:04:06 Unexpected Error - Cancelling MRP - MrpExp
22:04:08 Cancelling from MrpExp stopProcesses
This is my post from last week when we were struggling getting MRP to run different process sets on different schedules - due to a known bug with how the day of the week is evaluated. This has all stemmed from our nightly MRP process growing in length due to large order backlog and extending forecast as a result of the supply chain challenges the world is facing.
Not certain if this will prove to be connected, but symptomatically we had a similar issue.
Since go-live in 2018 we expected MRP to take 5-6 hours. We started digging into it because the PO suggestions in Epicor made no sense. As far as we could tell all settings were correct and reflected reality, but with overlapping inbound shipments and our outbound:inbound ratio being 1-day:6-weeks, we thought we were just asking too much of the system.
We set up a test environment and ran MRP filtered for one single part, over and over and over, to figure out exactly what each setting did.
In the end, we found the hierarchy of settings was killing us. If a given setting exists as a default in (for example) PartClass it only applies if the same setting in Part, PartPlant, Part Warehouse, Supplier, Supplier Price List, etc. is blank. For every setting that exists in more than on level, MRP seems to hunt up and down the hierarchy to determine which level takes precedence.
Now we run full regenerative every night. It takes 40 minutes, and the PO suggestions are about 85% accurate with analysts’ gut feeling and conversations with vendors making up the difference.
If the recent failures have consistently been since Sunday (Every night) check your Task schedules maybe this is related to the time change.
“For the past several days our nightly run of ‘Process MRP’ has failed”
Epicor will roll your schedules back an hour on the time change.
A couple months back we were getting a timeout error about once per week. We discovered the timeout setting in web.config and increased it to 10 hrs, and that address that particular error. The 2-hr trigger for ‘PROGRAM CANCELLED’ must be due to something else.
I would consider that but the first failure in this series occurred Fri eve and end of DST was early morning hours of Sun, so I’m thinking it’s something else.
Any updates install on app or SQL servers at that time? I know I got bit by that many times; even the slightest restart of .NET (I think) would do it. This is way outside my comfort zone, so hopefully this makes sense…
I think you might be on to something Jason. When we first started messing around with MRP settings, it was because we were hitting a timeout at 8 hours. The issue appeared seemingly out of the blue. But once we figured out what was going on, it made sense. We had slowly been generating more and more jobs over time due to backlog and so on. One day, we started bumping up against the limit.
In this case, when I review the log, what I see is that there must be “operations” within the process… and it appears to me that those operations have a timeout of 1 hour. We’ve been seeing it timeout at 1 hour when it fails in full regen mode, but last night’s failed at 2 hours. I believe that’s because the first operation completed in just under 1 hour, therefore did not timeout (see screenshots below)… and the process moved on to the next operation. I don’t know if I am using the right terminology, but I see the word “operation” in the error.
Does anyone know where there might be a timeout setting that controls these “operation” timeouts? I checked web.config and I don’t see one that looks like it would control this.
Another good question would be why does deleting the TFOrdDtl table take close to an hour. I found this in the archives. It’s over my head, though. @tkueppers maybe you can make sense of it?
There is an MRPTImeout setting that can be placed in the web.config. That solved a lot of our timeout errors while deleting unfirm jobs. If you look at the MRP log it tells you what those settings are at the top (at least in 10.2.400.13 it did)
aha! The query timeout is set to 3600 (60 seconds x 60 minutes). It’s failing to execute the delete query and then timing out. I bet this will get us past that timeout problem. I still want to investigate why it is taking an hour to execute… as it should only take seconds… like the other tables.
So, our MRP run generates (and deletes) over 1.5 million rows of JobMtl. It’s takes hours (by my math) to delete all of that but our timeout is also 3600, and when I read the “Delete of Table: JobMtl” line in the log, it supposedly took only 4 minutes. (Or zero; I don’t exactly know what “after” means there.) Details below if you expand it.
I’m not saying the timeout isn’t worth changing, since it helped @Paul_Millsaps . But I guess all I am saying is I have even less of a clue of what is going on than when this started.
My latest MRP log
Sunday, November 7, 2021 08:00:22
08:00:22 MRP Regeneration process begin.
08:00:22 ------------------------------------------------------------
08:00:22 Part Run MRP -> False
08:00:22 Cut Off Date -> 11/7/2031 12:00:00 AM
08:00:22 Schedule Start Date -> 11/7/2021 12:00:00 AM
08:00:22 Run Finite Scheduling -> False
08:00:22 Ignore Constrained Materials -> True
08:00:22 Allow Historical Dates -> True
08:00:22 Use Production Preparation Buffer -> False
08:00:22 Sort Level 0 MRP Jobs by Date -> False
08:00:22 Recycle MRP Jobs -> False
08:00:22 Get Details From Quote -> False
08:00:22 Run Multi Level Pegging -> False
08:00:22 Include Contract PO Parts -> False
08:00:22 Generate Purchase Schedules -> False
08:00:22 Number of MRP Processes -> 7
08:00:22 Number of Schedulers -> 7
08:00:22 Rough Cut Schedule When Getting Details -> True
08:00:22 Site List -> MfgSys
08:00:22 Part List ->
08:00:22 Part Class List ->
08:00:22 Product Group List ->
08:00:22 ------------------------------------------------------------
08:00:22 MRPRowCount (rows) -> 2000
08:00:22 MRPCommandTimeout (seconds) -> 3600
08:00:22 ------------------------------------------------------------
08:00:22 Process Control
08:00:22 Name -> ProcessorMain; Code -> Erp.Internal.MR.MrpExp.dll
08:00:22 Process Stop Queues -> Part~Delete~LotJobSplit~SchedJob~FirmJob~SaveLoad~QuotePart
08:00:22 Controller -> True; Range Start -> 0; Range End -> 0
08:00:22 Version 1.00.01
08:00:22 Name -> ProcessorMRP; Code -> Erp.Internal.MR.MrpExpCD.dll
08:00:22 Process Stop Queues ->
08:00:22 Controller -> False; Range Start -> 1; Range End -> 100
08:00:22 Version 1.00.01
08:00:22 Name -> ProcessorSched; Code -> Erp.Internal.MR.MrpExpSched.dll
08:00:22 Process Stop Queues ->
08:00:22 Controller -> False; Range Start -> 101; Range End -> 200
08:00:22 Version 1.00.01
08:00:22 ------------------------------------------------------------
08:00:22 Process Queue
08:00:22 Name -> BuildDelPOQueue; Queue In -> ; Queue Out -> DeleteAllPO; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> ; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> BuildDelJob; Queue In -> ; Queue Out -> DeleteJob; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> ; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> BuildDelTO; Queue In -> ; Queue Out -> DeleteTO; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> Delete; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> ProcessDelPO; Queue In -> DeleteAllPO; Queue Out -> ; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> ; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> ProcessDelJob; Queue In -> DeleteJob; Queue Out -> ; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> ; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> ProcessDelTO; Queue In -> DeleteTO; Queue Out -> ; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Delete; Check Group -> ; Repeat -> False; Regen Only -> True; Finite -> False
08:00:22 Name -> BuildSched; Queue In -> ; Queue Out -> PreJobF; Controller -> True
08:00:22 Alt Queue Out 1 -> PreJobI; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Sched; Check Group -> Sched; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> SchedJobF; Queue In -> PreJobF; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Sched; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> True
08:00:22 Name -> SchedJobI; Queue In -> PreJobI; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Sched; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> BuildQuoteJobQueue; Queue In -> ; Queue Out -> QuotePart; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Quote; Check Group -> Quote; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> ProcessQuoteJob; Queue In -> QuotePart; Queue Out -> ; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Quote; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> BuildNonPart; Queue In -> ; Queue Out -> Part; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> NonPart; Check Group -> NonPart; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> ProcessNonPart; Queue In -> Part; Queue Out -> SchedJobF; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> SchedJobI; Alt Queue Out 3 -> FirmJob
08:00:22 Group -> NonPart; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> LotJobSplitNonPart; Queue In -> LotJobSplit; Queue Out -> SchedJobF; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> SchedJobI; Alt Queue Out 3 -> FirmJob
08:00:22 Group -> NonPart; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> SchedNonPartF; Queue In -> SchedJobF; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> NonPart; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> True
08:00:22 Name -> SchedNonPartI; Queue In -> SchedJobI; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> NonPart; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> FirmJobNonPart; Queue In -> FirmJob; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> NonPart; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> BuildPart0; Queue In -> ; Queue Out -> Part; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Zero; Check Group -> Zero; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> ProcessPart0; Queue In -> Part; Queue Out -> ZeroJob; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Zero; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> LotJobSplitPart0; Queue In -> LotJobSplit; Queue Out -> ZeroJob; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Zero; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> SortPart0; Queue In -> ZeroJob; Queue Out -> SchedJobF; Controller -> True
08:00:22 Alt Queue Out 1 -> SchedJobI; Alt Queue Out 2 -> FirmJob; Alt Queue Out 3 ->
08:00:22 Group -> ZeroSort; Check Group -> ZeroSort; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> SchedPart0F; Queue In -> SchedJobF; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> ZeroSort; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> True
08:00:22 Name -> SchedPart0I; Queue In -> SchedJobI; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> ZeroSort; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> FirmJobPart0; Queue In -> FirmJob; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> ZeroSort; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> BuildPart; Queue In -> ; Queue Out -> Part; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Part; Check Group -> Part; Repeat -> True; Regen Only -> False; Finite -> False
08:00:22 Name -> ProcessPartL; Queue In -> Part; Queue Out -> SchedJobF; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> SchedJobI; Alt Queue Out 3 -> FirmJob
08:00:22 Group -> Part; Check Group -> ; Repeat -> True; Regen Only -> False; Finite -> False
08:00:22 Name -> LotJobSplitPartL; Queue In -> LotJobSplit; Queue Out -> SchedJobF; Controller -> False
08:00:22 Alt Queue Out 1 -> LotJobSplit; Alt Queue Out 2 -> SchedJobI; Alt Queue Out 3 -> FirmJob
08:00:22 Group -> Part; Check Group -> ; Repeat -> True; Regen Only -> False; Finite -> False
08:00:22 Name -> SchedPartLF; Queue In -> SchedJobF; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Part; Check Group -> ; Repeat -> True; Regen Only -> False; Finite -> True
08:00:22 Name -> SchedPartLI; Queue In -> SchedJobI; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> FirmJob; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Part; Check Group -> ; Repeat -> True; Regen Only -> False; Finite -> False
08:00:22 Name -> FirmJobPartL; Queue In -> FirmJob; Queue Out -> SaveLoad; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Part; Check Group -> ; Repeat -> True; Regen Only -> False; Finite -> False
08:00:22 Name -> ProcessLoad; Queue In -> ; Queue Out -> ; Controller -> True
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Load; Check Group -> Load; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 Name -> SaveLoad; Queue In -> SaveLoad; Queue Out -> ; Controller -> False
08:00:22 Alt Queue Out 1 -> ; Alt Queue Out 2 -> ; Alt Queue Out 3 ->
08:00:22 Group -> Load; Check Group -> ; Repeat -> False; Regen Only -> False; Finite -> False
08:00:22 ------------------------------------------------------------
08:00:22 Beginning of Delete Unfirm Jobs
09:40:18 Site: MfgSys After Delete of Table: ShopLoad
09:40:40 Site: MfgSys After Backout for JobMtl
09:40:45 Site: MfgSys After BackOut of JobAsmbl
09:42:03 Site: MfgSys After Delete of Table: JobProd
09:42:04 Site: MfgSys After Delete of Table: JobAudit
09:42:04 Site: MfgSys After Delete of Table: PcValueGrp
09:42:04 Site: MfgSys After Delete of Table: PcValueHead
09:42:04 Site: MfgSys After Delete of Table: PcValueSet
09:42:04 Site: MfgSys After Delete of Table: JobMtlRestriction
09:42:04 Site: MfgSys After Delete of Table: JObMtlRefDes
09:42:04 Site: MfgSys After Delete of Table: JobMtlInsp
09:42:04 Site: MfgSys After Delete of Table: JobMtlRestrictSubst
09:42:04 Site: MfgSys After Delete of Table: PcValueGrp
09:42:04 Site: MfgSys After Delete of Table: PcValueHead
09:42:04 Site: MfgSys After Delete of Table: PcValueSet
09:46:09 Site: MfgSys After Delete of Table: JobMtl
09:46:09 Site: MfgSys After Delete of Table: JobOperInsp
09:47:33 Site: MfgSys After Delete of Table: JobOper
09:49:47 Site: MfgSys After Delete of Table: JobOpDtl
09:49:47 Site: MfgSys After Delete of Table: JobOperRestriction
09:49:47 Site: MfgSys After Delete of Table: JobOperRestrictSubst
09:49:47 Site: MfgSys After Delete of Table: PartSug
09:49:47 Site: MfgSys After Delete of Table: PcValueGrp
09:49:47 Site: MfgSys After Delete of Table: PcValueHead
09:49:48 Site: MfgSys After Delete of Table: PcValueSet
09:50:28 Site: MfgSys After Delete of Table: JobAsmbl
09:50:28 Site: MfgSys After Delete of Table: JobAsmblRestriction
09:50:28 Site: MfgSys After Delete of Table: JobAsmblInsp
09:50:28 Site: MfgSys After Delete of Table: JobAsmblRestrictSubst
10:13:27 Site: MfgSys After Delete of Table: TFOrdDtl
10:13:28 Site: MfgSys After Delete of Table: JobPart
10:24:02 Site: MfgSys After Delete of Table: PartDtl
10:24:02 Site: MfgSys After Delete of Table: PegDmdMst
10:24:03 Site: MfgSys After Delete of Table: ChgLog
10:24:03 Site: MfgSys After Delete of Table: Memo
10:24:03 Site: MfgSys After Delete of Table: XFileAttch
10:24:03 Site: MfgSys After Delete of Table: ResourceTimeUsedSub
10:25:24 Site: MfgSys After Delete of Table: ResourceTimeUsed
10:25:24 Site: MfgSys After Delete of Table: JobOperMachParam
10:25:29 Site: MfgSys After Delete of Table: JobHead
10:25:29 End of Delete Unfirm Jobs===
10:25:29 Starting sub-processes
10:25:29 Deleting old Suggestions.
10:25:29 Deleting unfirm jobs...
10:25:30 Deleting transfer order suggestions ...
10:25:50 Scheduling Jobs...
10:25:58 Building Non-Part List
10:26:10 Building PartList Level: 0-0
21:00:21 Building PartList Level: 1-0
21:00:46 Building PartList Level: 2-0
21:25:25 Building PartList Level: 3-0
21:30:15 Building PartList Level: 4-0
21:30:42 Building PartList Level: 5-0
21:31:14 Building PartList Level: 6-0
21:31:26 Building PartList Level: 7-0
21:31:31 Processing Schedule Load
21:34:51 Processing Orphan PO's...
21:34:56 Processing Orphan PO:326806 Line:1 Rel:1.
21:35:11 MRP Regeneration process finished.
It’s nearly 2 years later and we still struggle with this same issue. The timeout setting has enabled us to keep MRP running even when it is terribly slow… We have bumped the MRP start time back to 7pm and it still sometimes runs over 12 hours and into the next workday. When it happens, it is 100% because of the gremlin that locks up TFOrdDtl table delete. It will take close to 2 hours per site when it’s really acting up. So I’ve seen it take over 6 hours to finish the deletes and start to rebuild the jobs. We do full regen each night and we like it that way. I know it can go faster because for several months at a time, it will seemingly hum along with only seconds to complete that table delete step. What on earth could cause it to fluctuate from a few seconds to a few hours?? I’m beginning to wonder if there’s bad data somewhere. An orphaned record? What could it be? Any ideas on what I could do to try to catch the culprit? As an example, two nights ago MRP completed at 3:14am, a respectable 8ish hours. I’ve seen it finish as early as 1:30-2:00am. But then last night it didn’t finish until 7:45am this morning.
Remember Spartan Motors - their BOMs for a fire truck are huge.
