OnBase Workflow Server Service Issues

[Eugene_Stasenko]

We've been having an issue with our daily summary lists not being sent out properly.  We have a timer setup to send out invoice approval lists every morning and usually this works fine... however lately we've had days where it hasn't worked.  I've discovered that restarting the OnBase Workflow Server service along with the Hyland Timer Service usually resolves the issue.  Looking at the event manager, I see two errors occurring that I suspect are tied into this:

ERROR: Error #13: Type mismatch: '[string: ""]' Line: 138 Source: Microsoft VBScript runtime error

and

Message box contents: Caption: Database Text: Fatal DB error: unable to connect to ODBC source! Available buttons: OK Automatic Response: OK

If anyone could provide any guidance, it would be very much appreciated.  On a separate note, we've tried restarting our server and have upgraded to the 13 client only to see that the same issue remained.

[Paul_Lloyd]

It sounds like the Workflow Timer Service is losing connection to the database.

This may be as a result of a network glitch, or even because of scheduled maintenance

I have given more details in my comments on this thread:

https://www.hyland.com/community/onbase_product_communities/business_process_automation_modules/work...

[Eugene_Stasenko]

Thanks!  I appreciate the input.  We just had this issue re-occur this morning and this time neither of those two errors poppped up.. instead, I see this one appearing over and over:

Volume Shadow Copy Service error: Unexpected error calling routine ConvertStringSidToSid(S-1-5-21-3309674670-2383054370-2238226778-2626.bak).  hr = 0x80070539, The security ID structure is invalid.
.

Operation:
   OnIdentify event
   Gathering Writer Data

Context:
   Execution Context: Shadow Copy Optimization Writer
   Writer Class Id: {4dc3bdd4-ab48-4d07-adb0-3bee2926fd7f}
   Writer Name: Shadow Copy Optimization Writer
   Writer Instance ID: {220e2802-fbb4-498f-9edc-31935f7d4985}

I want to say this appeared in our logs about 10 times prior to the report failing to go out again.