When you try to fail over a Microsoft SQL Server cluster group to a cluster node, SQL Server does not come online.
The SQL Server process does not generate any new SQL error log entry. However, the following entry is logged in the event log of the failover target cluster node:
Log Name: Application
Source: MSSQL$<SQL Cluster group name>
Event ID: 17141
Task Category: (2)
Level: Error
Keywords: Classic
Description:
Could not register Service Control Handler. Operating system error = 2310 (This shared resource does not exist.).
Additionally, you may notice a recorded Svchost.exe crash in the event log:
Log Name: Application
Source: Application Error
Event ID: 1000
Task Category: Application Crashing Events
Level: Error
Keywords: Classic
Description:
Faulting application name: svchost.exe_<serviceName>, version: <version>, time stamp: <time stamp>
Faulting module name: <module name>, version: <version#>, time stamp: <time stamp>
Exception code: <code>
Fault offset: <number>
Faulting process id: <PID>
Faulting application start time: <time stamp>
Faulting application path: C:\Windows\system32\svchost.exe
Faulting module path: <module full path>
Report Id: <GUID>
Faulting package full name:
Faulting package-relative application ID:
The crash event may be logged several hours or even several days before the SQL Server failover failure. Also, the entry may not specifically indicate the LanManServer service. The crash might be related to other services that share the same Svchost.exe instance with LanManServer.