Event ID 6237: Failover Cluster WMI Provider detected an invalid character
- Hi,
Four node Hyper-V cluster - originally Server 2008 RTM, upgraded to Server 2008 SP2 and then to Server 2008 R2.
I am recieving the following errors every 4 hours exactly and can't find any solution.
I did find http://support.microsoft.com/kb/958807 but this relates to 2008RTM and notes that the bug was fixed in SP2.
Any help would be appreciated.
Thanks
Log Name: Microsoft-Windows-FailoverClustering-WMIProvider/Admin
Source: Microsoft-Windows-FailoverClustering-WMIProvider
Date: 30/11/2009 08:10:29
Event ID: 6237
Task Category: Failover Cluster WMI Provider
Level: Error
Keywords:
User: RICHMOND\HO-VM-SCVMM01$
Computer: HO-BL01.trfs.org.uk
Description:
Failover Cluster WMI Provider detected an invalid character. The private property name 'ShutdownTimeoutInMinutes' had an invalid character but the provider failed to change it to a valid property name. Property names must start with A-Z or a-z, and valid characters for WMI property names are A-Z, a-z, 0-9, and '_'.
Event Xml:
<Event xmlns="http://schemas.microsoft.com/win/2004/08/events/event">
<System>
<Provider Name="Microsoft-Windows-FailoverClustering-WMIProvider" Guid="{0461BE3C-BC15-4BAD-9A9E-51F3FADFEC75}" />
<EventID>6237</EventID>
<Version>0</Version>
<Level>2</Level>
<Task>1</Task>
<Opcode>0</Opcode>
<Keywords>0x8000000000000000</Keywords>
<TimeCreated SystemTime="2009-11-30T08:10:29.434430200Z" />
<EventRecordID>151977</EventRecordID>
<Correlation />
<Execution ProcessID="6452" ThreadID="6884" />
<Channel>Microsoft-Windows-FailoverClustering-WMIProvider/Admin</Channel>
<Computer>HO-BL01.trfs.org.uk</Computer>
<Security UserID="S-1-5-21-2066427151-189991117-1850952788-15501" />
</System>
<EventData>
<Data Name="Parameter1">ShutdownTimeoutInMinutes</Data>
</EventData>
</Event>
Billy
Answers
Never mind - The following hotfix stops the error happening even though the specific problem is not mentioned in the KB article:
http://support.microsoft.com/kb/974930 - "An application or service that queries information about a failover cluster by using the WMI provider may experience low performance or a time-out exception"
Billy- Marked As Answer byBilly Clark Monday, November 30, 2009 10:40 PM
All Replies
Never mind - The following hotfix stops the error happening even though the specific problem is not mentioned in the KB article:
http://support.microsoft.com/kb/974930 - "An application or service that queries information about a failover cluster by using the WMI provider may experience low performance or a time-out exception"
Billy- Marked As Answer byBilly Clark Monday, November 30, 2009 10:40 PM
- Hi Billy,
Can elaborate a little more on the steps you did when you say you did an "Upgrade" from Win2008 to Win2008 R2? Because we don't support Upgrades to R2 with Failover Clusters, and if you pop in the Windows DVD and hit Upgrade on a server that is an active member of a cluster you probably noticed it blocks and throws an error saying it's unsupported. I'm more curious on what process you followed, and any other issues you've encountered.
Thanks!
Elden Hi Elden,
I have 2 x 4 node Hyper-V clusters currrently running around 40 x VM between them.
Both clusters were originally Server 2008 RTM with Hyper-V beta and were created very early in the Hyper-V time-line (we were part of virtulization TAP/RDP) and had been successively upgraded to SP2.
The clusters had been running in their current configuration for well over 12 months with Hyper-V being very reliable.
I upgraded the clusters to Server 2008 R2 by:
Migrating all VM to 2 nodes (resources available on only 2 nodes required that some VM were shut down)
Evicting the other 2 nodes
In-place upgrade including drivers etc of the evicted nodes
Creating a new Server 2008 R2 cluster on 2 nodes
Used System Centre Vitrual Machine Manager 2008 R2 to migrate the virtual machines to new cluster and different storage (primarily CSV but some single VM/LUN)
Destroyed the old cluster
Upgraded 2 x remaining nodes to Server 2008 R2 and joined them to the Server 2008 R2 cluster
I didn't have any great issues and don't use any special SAN hardware and luckily the server hardware (HP) had drivers available before the R2 release.
The whole process on both clusters was completed in 2 and a bit days one weekend in September.
Hope this helps.
Billy- Hi Elden & Billy,
I have a related problem, triggered by SCVMM 2008R2, and started the following thread:
http://social.technet.microsoft.com/Forums/en-US/virtualmachinemgrclustering/thread/ba6de1e6-9b79-4502-8978-a5436d4e99a5
M Fawzi suggested I had a look at your case, which was a good hint. To satisfy Elden's possible curiosity, the cluster was first set up in Windows 2008 DC with Hyper-V, and was later upgraded to Windows 2008 R2 DC. You refer to hotfix described in kb974930. I have no test system available that simulates my production environment entirely. Can Elden recommand I install the hotfix to remove the error?
Hope you can help,
Wim - Hi Wim
In my opinion there are a number of underlying WMI problems with the original Serrver2008 RTM, Hyper-V and SCVMM 2008 versions.
They dont appear to affect the stability just the inter-application reporting i.e. Hyper-V to SCVMM to SCOM
For example see this still unresolved issue: http://social.technet.microsoft.com/Forums/en-US/virtualmachinemgrclustering/thread/f7f1c4cc-1e26-421b-87d5-ef7d4472122b
This WMI issue is not evident when using a new VM on a clean installed Server 2008R2 with SCVMM2008R2 and all stats are reported correctly.
I installed http://support.microsoft.com/kb/974930 on 2 x 4 nodes clusters in a live production environment with no issues and all the Event ID: 6237 issues disappeared but this does not fix the other issue.
There was also a SCVMM hotfix rollup issued yesterday that you should also consider: http://support.microsoft.com/kb/978560
Again I have applied this to SCVMM and updated the agents on 2 x 4 node clusters in production with no issues.
Good luck
Billy