SQL Server Agent doesn't start after upgrading to build 3239

Here is the KB article outlining the problem and two work-arounds
http://support.microsoft.com/kb/943525

BTW, the log entry is being analyzed from a maching that did not initially
create it, possibly from a different cluster node. The missing event
information is not the underlying error, it is simply a description when the
event data comes from a different system and the local one does not have the
same component to decode the event.


--
Geoff N. Hiten
Senior SQL Infrastructure Consultant
Microsoft SQL Server MVP


"RQ" wrote in message
news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
>I have an active/active cluster running under SQL Server 2005 64 bit sp2.
> After upgrading to build 3239 the SQL Server Agent won't start anymore...
> The
> error captured in the log file is:
>
> SQLServer Error: 22022, CryptUnprotectData() returned error -2146893813,
> 'Key not valid for use in specified state.' [SQLSTATE 42000]
>
> And the log captured in the even log is:
>
> The description for Event ID ( 17052 ) in Source ( MSSQLSERVER ) cannot be
> found. The local computer may not have the necessary registry information
> or
> message DLL files to display messages from a remote computer. You may be
> able
> to use the /AUXSOURCE= flag to retrieve this description; see Help and
> Support for details. The following information is part of the event:
> SQLServerAgent Monitor failed to restart SQLServerAgent after
> SQLServerAgent
> terminated unexpectedly (reason: SQLSCMControl() returned error 5040, 'The
> cluster node already exists.')..
>
> Does anyone know how to resolve this?
>

Geoff:
Thank you very much for your quick response!!! I tried first workaround
and did not work. I would like to avoid the second work around since
delegation on the cluster was not setup before the upgrade and therefore all
services were operational. Is there another way to fix this?

Thank you very much for your help
RQ


"Geoff N. Hiten" wrote:

> Here is the KB article outlining the problem and two work-arounds
> http://support.microsoft.com/kb/943525
>
> BTW, the log entry is being analyzed from a maching that did not initially
> create it, possibly from a different cluster node. The missing event
> information is not the underlying error, it is simply a description when the
> event data comes from a different system and the local one does not have the
> same component to decode the event.
>
>
> --
> Geoff N. Hiten
> Senior SQL Infrastructure Consultant
> Microsoft SQL Server MVP
>
>
> "RQ" wrote in message
> news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
> >I have an active/active cluster running under SQL Server 2005 64 bit sp2.
> > After upgrading to build 3239 the SQL Server Agent won't start anymore...
> > The
> > error captured in the log file is:
> >
> > SQLServer Error: 22022, CryptUnprotectData() returned error -2146893813,
> > 'Key not valid for use in specified state.' [SQLSTATE 42000]
> >
> > And the log captured in the even log is:
> >
> > The description for Event ID ( 17052 ) in Source ( MSSQLSERVER ) cannot be
> > found. The local computer may not have the necessary registry information
> > or
> > message DLL files to display messages from a remote computer. You may be
> > able
> > to use the /AUXSOURCE= flag to retrieve this description; see Help and
> > Support for details. The following information is part of the event:
> > SQLServerAgent Monitor failed to restart SQLServerAgent after
> > SQLServerAgent
> > terminated unexpectedly (reason: SQLSCMControl() returned error 5040, 'The
> > cluster node already exists.')..
> >
> > Does anyone know how to resolve this?
> >
>
>

Did you put the new service accounts into the cluster security groups you
used when you built the cluster?

--
Geoff N. Hiten
Senior SQL Infrastructure Consultant
Microsoft SQL Server MVP



"RQ" wrote in message
news:C170DEBF-0374-4110-A640-018DE9AE26F5@microsoft.com...
> Geoff:
> Thank you very much for your quick response!!! I tried first
> workaround
> and did not work. I would like to avoid the second work around since
> delegation on the cluster was not setup before the upgrade and therefore
> all
> services were operational. Is there another way to fix this?
>
> Thank you very much for your help
> RQ
>
>
> "Geoff N. Hiten" wrote:
>
>> Here is the KB article outlining the problem and two work-arounds
>> http://support.microsoft.com/kb/943525
>>
>> BTW, the log entry is being analyzed from a maching that did not
>> initially
>> create it, possibly from a different cluster node. The missing event
>> information is not the underlying error, it is simply a description when
>> the
>> event data comes from a different system and the local one does not have
>> the
>> same component to decode the event.
>>
>>
>> --
>> Geoff N. Hiten
>> Senior SQL Infrastructure Consultant
>> Microsoft SQL Server MVP
>>
>>
>> "RQ" wrote in message
>> news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
>> >I have an active/active cluster running under SQL Server 2005 64 bit
>> >sp2.
>> > After upgrading to build 3239 the SQL Server Agent won't start
>> > anymore...
>> > The
>> > error captured in the log file is:
>> >
>> > SQLServer Error: 22022, CryptUnprotectData() returned
>> > error -2146893813,
>> > 'Key not valid for use in specified state.' [SQLSTATE 42000]
>> >
>> > And the log captured in the even log is:
>> >
>> > The description for Event ID ( 17052 ) in Source ( MSSQLSERVER ) cannot
>> > be
>> > found. The local computer may not have the necessary registry
>> > information
>> > or
>> > message DLL files to display messages from a remote computer. You may
>> > be
>> > able
>> > to use the /AUXSOURCE= flag to retrieve this description; see Help and
>> > Support for details. The following information is part of the event:
>> > SQLServerAgent Monitor failed to restart SQLServerAgent after
>> > SQLServerAgent
>> > terminated unexpectedly (reason: SQLSCMControl() returned error 5040,
>> > 'The
>> > cluster node already exists.')..
>> >
>> > Does anyone know how to resolve this?
>> >
>>
>>

I used the same domain account I was using before which is not a domain
account... Just regular account that has admin access to both nodes of my
cluster and that belongs to cluster security groups I created when setting up
the cluster for the first time Sorry to keep bothering you but do you have
any other ideas?

"Geoff N. Hiten" wrote:

> Did you put the new service accounts into the cluster security groups you
> used when you built the cluster?
>
> --
> Geoff N. Hiten
> Senior SQL Infrastructure Consultant
> Microsoft SQL Server MVP
>
>
>
> "RQ" wrote in message
> news:C170DEBF-0374-4110-A640-018DE9AE26F5@microsoft.com...
> > Geoff:
> > Thank you very much for your quick response!!! I tried first
> > workaround
> > and did not work. I would like to avoid the second work around since
> > delegation on the cluster was not setup before the upgrade and therefore
> > all
> > services were operational. Is there another way to fix this?
> >
> > Thank you very much for your help
> > RQ
> >
> >
> > "Geoff N. Hiten" wrote:
> >
> >> Here is the KB article outlining the problem and two work-arounds
> >> http://support.microsoft.com/kb/943525
> >>
> >> BTW, the log entry is being analyzed from a maching that did not
> >> initially
> >> create it, possibly from a different cluster node. The missing event
> >> information is not the underlying error, it is simply a description when
> >> the
> >> event data comes from a different system and the local one does not have
> >> the
> >> same component to decode the event.
> >>
> >>
> >> --
> >> Geoff N. Hiten
> >> Senior SQL Infrastructure Consultant
> >> Microsoft SQL Server MVP
> >>
> >>
> >> "RQ" wrote in message
> >> news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
> >> >I have an active/active cluster running under SQL Server 2005 64 bit
> >> >sp2.
> >> > After upgrading to build 3239 the SQL Server Agent won't start
> >> > anymore...
> >> > The
> >> > error captured in the log file is:
> >> >
> >> > SQLServer Error: 22022, CryptUnprotectData() returned
> >> > error -2146893813,
> >> > 'Key not valid for use in specified state.' [SQLSTATE 42000]
> >> >
> >> > And the log captured in the even log is:
> >> >
> >> > The description for Event ID ( 17052 ) in Source ( MSSQLSERVER ) cannot
> >> > be
> >> > found. The local computer may not have the necessary registry
> >> > information
> >> > or
> >> > message DLL files to display messages from a remote computer. You may
> >> > be
> >> > able
> >> > to use the /AUXSOURCE= flag to retrieve this description; see Help and
> >> > Support for details. The following information is part of the event:
> >> > SQLServerAgent Monitor failed to restart SQLServerAgent after
> >> > SQLServerAgent
> >> > terminated unexpectedly (reason: SQLSCMControl() returned error 5040,
> >> > 'The
> >> > cluster node already exists.')..
> >> >
> >> > Does anyone know how to resolve this?
> >> >
> >>
> >>
>
>

oops correction:
I used the same domain account I was using before which is not an Admin
domain
> account... Just regular account that has admin access to both nodes of my
> cluster and that belongs to cluster security groups I created when setting up
> the cluster for the first time Sorry to keep bothering you but do you have
> any other ideas? Also, I was able to see connectivity via SQL Server Profiler but it logs out...

"RQ" wrote:

> I used the same domain account I was using before which is not a domain
> account... Just regular account that has admin access to both nodes of my
> cluster and that belongs to cluster security groups I created when setting up
> the cluster for the first time Sorry to keep bothering you but do you have
> any other ideas?
>
> "Geoff N. Hiten" wrote:
>
> > Did you put the new service accounts into the cluster security groups you
> > used when you built the cluster?
> >
> > --
> > Geoff N. Hiten
> > Senior SQL Infrastructure Consultant
> > Microsoft SQL Server MVP
> >
> >
> >
> > "RQ" wrote in message
> > news:C170DEBF-0374-4110-A640-018DE9AE26F5@microsoft.com...
> > > Geoff:
> > > Thank you very much for your quick response!!! I tried first
> > > workaround
> > > and did not work. I would like to avoid the second work around since
> > > delegation on the cluster was not setup before the upgrade and therefore
> > > all
> > > services were operational. Is there another way to fix this?
> > >
> > > Thank you very much for your help
> > > RQ
> > >
> > >
> > > "Geoff N. Hiten" wrote:
> > >
> > >> Here is the KB article outlining the problem and two work-arounds
> > >> http://support.microsoft.com/kb/943525
> > >>
> > >> BTW, the log entry is being analyzed from a maching that did not
> > >> initially
> > >> create it, possibly from a different cluster node. The missing event
> > >> information is not the underlying error, it is simply a description when
> > >> the
> > >> event data comes from a different system and the local one does not have
> > >> the
> > >> same component to decode the event.
> > >>
> > >>
> > >> --
> > >> Geoff N. Hiten
> > >> Senior SQL Infrastructure Consultant
> > >> Microsoft SQL Server MVP
> > >>
> > >>
> > >> "RQ" wrote in message
> > >> news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
> > >> >I have an active/active cluster running under SQL Server 2005 64 bit
> > >> >sp2.
> > >> > After upgrading to build 3239 the SQL Server Agent won't start
> > >> > anymore...
> > >> > The
> > >> > error captured in the log file is:
> > >> >
> > >> > SQLServer Error: 22022, CryptUnprotectData() returned
> > >> > error -2146893813,
> > >> > 'Key not valid for use in specified state.' [SQLSTATE 42000]
> > >> >
> > >> > And the log captured in the even log is:
> > >> >
> > >> > The description for Event ID ( 17052 ) in Source ( MSSQLSERVER ) cannot
> > >> > be
> > >> > found. The local computer may not have the necessary registry
> > >> > information
> > >> > or
> > >> > message DLL files to display messages from a remote computer. You may
> > >> > be
> > >> > able
> > >> > to use the /AUXSOURCE= flag to retrieve this description; see Help and
> > >> > Support for details. The following information is part of the event:
> > >> > SQLServerAgent Monitor failed to restart SQLServerAgent after
> > >> > SQLServerAgent
> > >> > terminated unexpectedly (reason: SQLSCMControl() returned error 5040,
> > >> > 'The
> > >> > cluster node already exists.')..
> > >> >
> > >> > Does anyone know how to resolve this?
> > >> >
> > >>
> > >>
> >
> >

The service account does not have to be and should not be a local admin on
the nodes. Simply putting it into the groups sets all the permissions
necessary. Try removing it from the local admins group on each node.

--
Geoff N. Hiten
Senior SQL Infrastructure Consultant
Microsoft SQL Server MVP




"RQ" wrote in message
news:5A4FD5B7-8FCC-4424-AFAE-8BDAF967FC7B@microsoft.com...
>I used the same domain account I was using before which is not a domain
> account... Just regular account that has admin access to both nodes of my
> cluster and that belongs to cluster security groups I created when setting
> up
> the cluster for the first time Sorry to keep bothering you but do you
> have
> any other ideas?
>
> "Geoff N. Hiten" wrote:
>
>> Did you put the new service accounts into the cluster security groups you
>> used when you built the cluster?
>>
>> --
>> Geoff N. Hiten
>> Senior SQL Infrastructure Consultant
>> Microsoft SQL Server MVP
>>
>>
>>
>> "RQ" wrote in message
>> news:C170DEBF-0374-4110-A640-018DE9AE26F5@microsoft.com...
>> > Geoff:
>> > Thank you very much for your quick response!!! I tried first
>> > workaround
>> > and did not work. I would like to avoid the second work around since
>> > delegation on the cluster was not setup before the upgrade and
>> > therefore
>> > all
>> > services were operational. Is there another way to fix this?
>> >
>> > Thank you very much for your help
>> > RQ
>> >
>> >
>> > "Geoff N. Hiten" wrote:
>> >
>> >> Here is the KB article outlining the problem and two work-arounds
>> >> http://support.microsoft.com/kb/943525
>> >>
>> >> BTW, the log entry is being analyzed from a maching that did not
>> >> initially
>> >> create it, possibly from a different cluster node. The missing event
>> >> information is not the underlying error, it is simply a description
>> >> when
>> >> the
>> >> event data comes from a different system and the local one does not
>> >> have
>> >> the
>> >> same component to decode the event.
>> >>
>> >>
>> >> --
>> >> Geoff N. Hiten
>> >> Senior SQL Infrastructure Consultant
>> >> Microsoft SQL Server MVP
>> >>
>> >>
>> >> "RQ" wrote in message
>> >> news:FE7C58CB-C728-44AC-9FDE-F6D732038D4C@microsoft.com...
>> >> >I have an active/active cluster running under SQL Server 2005 64 bit
>> >> >sp2.
>> >> > After upgrading to build 3239 the SQL Server Agent won't start
>> >> > anymore...
>> >> > The
>> >> > error captured in the log file is:
>> >> >
>> >> > SQLServer Error: 22022, CryptUnprotectData() returned
>> >> > error -2146893813,
>> >> > 'Key not valid for use in specified state.' [SQLSTATE 42000]
>> >> >
>> >> > And the log captured in the even log is:
>> >> >
>> >> > The description for Event ID ( 17052 ) in Source ( MSSQLSERVER )
>> >> > cannot
>> >> > be
>> >> > found. The local computer may not have the necessary registry
>> >> > information
>> >> > or
>> >> > message DLL files to display messages from a remote computer. You
>> >> > may
>> >> > be
>> >> > able
>> >> > to use the /AUXSOURCE= flag to retrieve this description; see Help
>> >> > and
>> >> > Support for details. The following information is part of the event:
>> >> > SQLServerAgent Monitor failed to restart SQLServerAgent after
>> >> > SQLServerAgent
>> >> > terminated unexpectedly (reason: SQLSCMControl() returned error
>> >> > 5040,
>> >> > 'The
>> >> > cluster node already exists.')..
>> >> >
>> >> > Does anyone know how to resolve this?
>> >> >
>> >>
>> >>
>>
>>