SMS_STATE_SYSTEM - SQL Message 6522 Severity 16

Harjit

I recently upgraded my SCCM server to 1511 with SQL Server 2012 R2 SP3 on Server 2008 R2. I'm seeing the following errors
in SMS_STATE_SYSTEM in Component Status. Any idea what the error refers to and the possible fixes? Thanks.

Microsoft SQL Server reported SQL message 6522, severity 16: [42000][6522][Microsoft][SQL Server Native Client 11.0][SQL Server]A .NET Framework error occurred during execution of
user-defined routine or aggregate "fnMDMCalculateHash":

System.Data.SqlTypes.SqlNullValueException: Data is Null. This method or prop

wally_mead

Hi Harjit. I believe we met in Minneapolis last week, did we not? If so, I hope that you enjoyed the conference. I seem to recall that this is a known issue (sounds really familiar). And is not really an issue, and something that you can safely ignore (assuming my memory is not failing me). Just to verify, do you see other state messages being processed fine? Such as for client upgrades, or deployment status, or patch scan or deployment status? if they all do process and get displayed appropriately (at least the results) in the Console, then all is good, and you can ignore that message. I want to say it was fixed in 1602. Let me know if other state messages are processed fine or not.

John_Fitzpatrick

When you upgraded did you get asked if you were working in online or offline mode? If you selected online mode and you don't have intune the following article sounds like it might be useful.

https://blogs.technet.microsoft.com/charlesa_us/2016/05/02/configmgr-current-branch-service-connection-point-and-whats-in-the-telemetry/

About half way down the page there is reference to "fnMDMCalculateHash":

Harjit

Yes, I did the online mode during install. The article is intriguing but it still doesn't provide any insight on a fix. Thanks for the reply.

[Deleted User]

Paging @wally_mead

wally_mead

Hi Harjit. I believe we met in Minneapolis last week, did we not? If so, I hope that you enjoyed the conference. I seem to recall that this is a known issue (sounds really familiar). And is not really an issue, and something that you can safely ignore (assuming my memory is not failing me). Just to verify, do you see other state messages being processed fine? Such as for client upgrades, or deployment status, or patch scan or deployment status? if they all do process and get displayed appropriately (at least the results) in the Console, then all is good, and you can ignore that message. I want to say it was fixed in 1602. Let me know if other state messages are processed fine or not.

Harjit

Hi Wally, yes we met several times last week at MMS. Remember the selfie? :-)
Yes, all other state messages seem to be fine with an occasional SMS_INVENTORY_DATA_LOADER error but all seems to be working as expected. I'll upgrade to 1602 from 1511 and see if that clears this error. Thanks for the response.

wally_mead

Yes, do remember your selfie ;-) Inventory Data Loader will often report errors from attempting to process hardware inventory. You should certainly validate that they are really issues, and that they do get resolved in future hardware inventory reports, but not uncommon at all.

Harjit

Harjit said:

I recently upgraded my SCCM server to 1511 with SQL Server 2012 R2 SP3 on Server 2008 R2. I'm seeing the following errors
in SMS_STATE_SYSTEM in Component Status. Any idea what the error refers to and the possible fixes? Thanks.

Microsoft SQL Server reported SQL message 6522, severity 16: [42000][6522][Microsoft][SQL Server Native Client 11.0][SQL Server]A .NET Framework error occurred during execution of
user-defined routine or aggregate "fnMDMCalculateHash":

System.Data.SqlTypes.SqlNullValueException: Data is Null. This method or prop

These errors are still prevalent in 1602. Doesn't seem to be fixed yet.

wally_mead

Sorry to hear that Harjit. It's possible that my memory is not what it used to be, but was pretty sure that it was to be resolved. Anyway, I fired off an email to a dev contact at Microsoft in the Configuration Manager product group to see what he recalls about it (the guy I want to say had reported it as a known, non-toxic, issue previously. I will update when I get a response from him.

Harjit

Thanks @wally_mead
I also have it posted in TechNet but no responses so far. Even MS support doesn't seem to have any answers at this time. I'm anxious to hear what you hear back. Thanks again for taking the time to look into this issue.

wally_mead

Wow, if you have posted in the TechNet forums with no answer, that is surprising. A lot of the MVPs monitor those (I don't much any longer) and stay on top of them. And if product support has not heard of it, maybe I'm thinking of something else. Will see what my dev contact says and report back.

wally_mead

Hi Harjit,

Just wanted to let you know that I've emailed the dev, and pinged him again, but still no answer. So I've not forgotten about this, just waiting for him to respond. I believe they are pretty busy working on a new build to be released, so that is likely consuming all his cycles.

Sorry for the delay. Did the posting in the TechNet forums result in anything for you yet?

Wally

Harjit

Wally,

Thanks for the update. I appreciate it. 
No, there is zero response on my TechNet post. It's disappointing especially when there are so many MVPs in this field. 

wally_mead

Sorry to hear that. I'll let you know when I do get a response from the Microsoft product group developer. But to confirm, everything else still seems to be working fine, correct?

Harjit

@wally_mead Sorry for the late response. I thought I had responded but overlooked it. Yes, everything else seems to be working fine. 

wally_mead

Glad to hear it. Still no response from the dev. I know that they are working on getting their next build ready, so assuming that he's pretty backlogged on that. I will respond when I hear anything back for you.

wally_mead

Hi Harjit,

I finally got a response back. Not what I was thinking, but give it a shot. The dev said:

For this particular issue, it is likely due to some bad state messages reported from clients. If possible, customer can double check the statesys.log and it will tell the CM Admins which particular .SMX file is causing the issue and it is likely they are moved to the corrupted folder (<cminstalldir>\ \inboxes\auth\statesys.box\corrupt)

If so, can they share the offending .SMX files and the statesys.log/lo_

Check that out and let me know if it resolves the issue. I've seen corrupt messages before, never in the State System however.

Wally

Harjit

Hey Wally,

Thank you very much for following up on this. I'll give this a try and provide some feedback. Thanks again.

wally_mead

Great, hope his thoughts help out.