How to disabel the Microsoft Search Service

Hi everybody,

I work with SQL Server 2000 SP2 in an Active/active Cluster.
Two days ago, one the sql server fulltext resource of one node just failed suddenly.
Now I am realizing that Search Engine Utility has not been used by any of my databases, so I would like do disable this service...
Does anybody know how can I get rid of the Microsoft Search Service.
I don't need this service active!
In Service Manager the Microsoft Search is in an unkown state.
In cluster Administrator the sql server full text resource is in a failed state.
Do I need to delete the full text resource or is there any other way to disable the Search Service?
Any help or advice will be very important to me...
Thanks,Search service is a shared service by all SQL Server instances.

Can you take off-line on SQL Server FullText resources from all SQL Server instances from cluster administrator?

You shouldn't remove the search service even though you currently do not use it. Try take it off-line.

Becuase your search service is in unknow status, I would move all cluster groups to the good node and restart the machine that search service is not function properly.|||Hi Allen_Cui

Thank you for your reply...
I have already tried to move all the groups to the good node and the result was:
Microsoft Search is in "stopped" state in Service Manager... -> that's ok!
SQL Server Full Text resource was in "failed" state in cluster Administrator... -> that's not ok!
But before I try the manual failover I selected the option "Do not restart" in the properties of the failed Full Text Resource...
My problem now is: I want to convert the state of the failed Full Text Resource to an offline state the is more pleasant to see...
Do you have any idea?
I will be waiting for your help...|||I would check the machine application/system logs to see any errors related FullText resource and search service.

How to detect BCP success/fail in batch file?

We are calling BCP from a batch file but cannot figure out how to detect
whether it was successful or if it failed. With osql we use EXIT() with a
value to indicate success or failure. Is there a way to do this with BCP? Or
does it report success or failure to the calling process or set an operating
system error value?
TIA
Michael MacGregor
Database Architect
Michael MacGregor (macnoknifespam@.noemailspam.com) writes:
> We are calling BCP from a batch file but cannot figure out how to detect
> whether it was successful or if it failed. With osql we use EXIT() with
> a value to indicate success or failure. Is there a way to do this with
> BCP? Or does it report success or failure to the calling process or set
> an operating system error value?
BCP does indeed set %ERRORLEVEL%. However, your and BCP's idea of what
is an error may not be the same. BCP will return with an error status
if for instance the data file cannot be found. It may also set a return
status if the end of file is in the middle of a record (which usually
means that your format specification is wrong). But I seem to recall
that if one more rows fail to import, that it does not set ERRORLEVEL -
not even if all records fail.
You can use the -e option to get error information into a file, this
captures problems with individual records.
Erland Sommarskog, SQL Server MVP, esquel@.sommarskog.se
Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx
|||Aha! When I check the ERRORLEVEL value, whether the BCP IN succeeds or fails
(failure caused by PK constraint), the value is the same, 0. However I
realised that I might not be using the latest version of BCP and checked, it
was BCP v8, so I updated my PATH to use BCP v9 and then it sets the
ERRORLEVEL correctly.
So I guess it's best to use the latest version.
Michael MacGregor
Database Architect
|||Michael MacGregor (macnoknifespam@.noemailspam.com) writes:
> Aha! When I check the ERRORLEVEL value, whether the BCP IN succeeds or
> fails (failure caused by PK constraint), the value is the same, 0.
> However I realised that I might not be using the latest version of BCP
> and checked, it was BCP v8, so I updated my PATH to use BCP v9 and then
> it sets the ERRORLEVEL correctly.
Interesting. That's seems to be an improvement.
However, I would test a lot more error situations to see if all set
ERRORLEVEL.
Erland Sommarskog, SQL Server MVP, esquel@.sommarskog.se
Books Online for SQL Server 2005 at
http://www.microsoft.com/technet/prodtechnol/sql/2005/downloads/books.mspx
Books Online for SQL Server 2000 at
http://www.microsoft.com/sql/prodinfo/previousversions/books.mspx
|||I've got round the immediate problem by using a SQL script incorporating
BULK INSERT. It does the job with the error handling I need.
Although I will take your advice under due consideration as I have another
process that needs to be "error aware" that also uses BCP right now, and
also is using v8.0. So I will have to update that to v9.0 and test it to
make sure it catches the types of errors that tend to occur.
Thanks.
Michael MacGregor
Database Architect

How to define token syntax in MSSQL2005 sp1?

There is one token in my Agent Job $WMI(DatabaseName)

Now,I defined this using $(ESCAPE_NONE(WMI(DatabaseName)))

but failed and prompted: Variable WMI(DatabaseName) not found

What should i do for this? thanks

From the updated books online:

For jobs that run in response to WMI alerts, the value of the property specified by property. For example, $(WMI(DatabaseName)) provides the value of the DatabaseName property for the WMI event that caused the alert to run.

So I believe you don't need the ESCAPE_NONE...

|||

If do not add ESCAPE_NONE,prompt "For SQL Server 2005 Service Pack 1 or later, all job steps with tokens must be updated with a macro before the job can run"

Discover In SQL Server 2005 SP1, the SQL Server Agent job step token syntax has changed

url：http://support.microsoft.com/kb/915845

I want to do something for each database just was created,So I could not specifiy the value of database, how to deal with that? thanks

|||

Perhaps the tokens remain disabled?

Because access to Eventlog is not always secured, the alerts are disabled by default. To get the substitutions to work, you should ensure that only members of trusted groups have write permissions to Eventlog, then enable these tokens on the Agent Properties Dialog Alert System tab, or you can set the AlertReplaceRuntimeTokens reg key.

jkh

How to define token syntax in MSSQL2005 sp1?

There is one token in my Agent Job $WMI(DatabaseName)

Now,I defined this using $(ESCAPE_NONE(WMI(DatabaseName)))

but failed and prompted: Variable WMI(DatabaseName) not found

What should i do for this? thanks

From the updated books online:

For jobs that run in response to WMI alerts, the value of the property specified by property. For example, $(WMI(DatabaseName)) provides the value of the DatabaseName property for the WMI event that caused the alert to run.

So I believe you don't need the ESCAPE_NONE...

|||

If do not add ESCAPE_NONE,prompt "For SQL Server 2005 Service Pack 1 or later, all job steps with tokens must be updated with a macro before the job can run"

Discover In SQL Server 2005 SP1, the SQL Server Agent job step token syntax has changed

url：http://support.microsoft.com/kb/915845

I want to do something for each database just was created,So I could not specifiy the value of database, how to deal with that? thanks

|||

Perhaps the tokens remain disabled?

Because access to Eventlog is not always secured, the alerts are disabled by default. To get the substitutions to work, you should ensure that only members of trusted groups have write permissions to Eventlog, then enable these tokens on the Agent Properties Dialog Alert System tab, or you can set the AlertReplaceRuntimeTokens reg key.

jkh