Hi, Im having issues with the Transform.Common job failing in my SCSM 2012. This did run ok until early July and has failed since.

When I run a get-scdwjob it shows the job start date of 10/7/2012 which is last month and a finish date of 16/08/2012 (today). Its almost like the job has never completed even though it fails every run. Also the BatchID of the transform.common job has not changed in quite some time.

The module which is failing is 1306 TransformEntityRelatesToEntityFact, if I run a get-scdwjobmodule and pipe the output to a file this is part of what I get (theres pages showing every failed run)-

"</Error><Error EventTime="2012-08-15T16:32:28.5406785Z">ErrorNumber="50000" Message="Unable to acquire applock - another instance of the module must already be running." Severity="18" State="1" ProcedureName="InitializeTransform" LineNumber="52" Task="(null)"</Error></Errors>

ModuleTypeName      : System

ModuleName          : TransformEntityRelatesToEntityFact

ModuleDescription   : Deployment Execution Step

JobName             : Transform.Common

CategoryName        : Transform

Description         : Transform.Common

CreationTime        : 15/08/2012 16:32:28

The same error from Operations Manager Event Log-

ETL Module Execution failed:

ETL process type: Transform

Batch ID: 5954

Module name: TransformEntityRelatesToEntityFact

Message: ErrorNumber="50000" Message="Unable to acquire applock - another instance of the module must already be running." Severity="18" State="1" ProcedureName="InitializeTransform" LineNumber="52" Task="(null)"

So, it seems the transform.common job thinks theres a TransformEntityRelatesToEntityFact module already running which is causing the current run of the job to fail. Ive restarted the services, Ive disabled all DW jobs and only ran tansform.common, Ive extended the period in-between runs of transform.common and I still get this error every time it runs. The job does create a temporary table in the DWRepository DB which Ive also tried dropping then running the task and it still fails.

Any ideas?

Cheers,

Max.

• Edited by MaxUK Thursday, August 16, 2012 10:27 AM

This is now fixed, I ran the get-scdwwatermark command against the failing module to find which other modules, etc were attached to it. One of these showed a date watermark of mid July - the same date and time which showed as the job start time when I ran a get-scdwjob. I reset this to a day later using set-scdwwatermark and started the job running which succeeded ok. So, transform.common is now working and we can run reports against the once missing data. I can only assume that something funny must have happened to the mid July run which caused the job to fail but not quite end, resetting the watermark fixed this.

Cheers,

Max.

• Marked as answer by MaxUK Wednesday, August 22, 2012 9:16 AM

I have been told that your fix will cause DW to lose all data from the time when the job failed until the new watermark time.

• Proposed as answer by CCOTS-KN Monday, August 18, 2014 4:19 PM

Make sure you apply the latest update rollup