In Rodney R. Fournier's blog it states:
"If MSDTC is used with SQL, you should ensure it is in it's own group,
on it's own disk resource. Lastly, Microsoft recommends to run the
MSDTC & SQL group on the same node, and that the MSDTC is brought
online first when the groups are moved."
("http://msmvps.com/clustering/archive/2004/10/25/16672.aspx").
I have a couple of questions about this but let first list my
assumptions.
Best:
For high performance, installing an MSDTC Resource to it in its own
group. This group should have its own disk and IP resources.
Good:
The next best method is to put the MSDTC resource in the cluster or
quorum group.
Worse:
Put the MSDTC resource in SQL Group.
What are the reasons for running the MSDTC & SQL group on the same
node, beyond these potential Win2003 cluster issues listed in:
http://support.microsoft.com/default...b;en-us;889706
http://support.microsoft.com/kb/821751/
?
If DTC is in the quorum group is it still best to put SQL and quorum
group on the same node?
Our system admin is religious about putting the SQL group on a
different node from the Quorum & DTC groups. Is there a best practice
here or a KB I can point to?
Are there any implications not having SQL dependent of MSDTC starting
first (MSDTC is brought online first when the groups are moved)
beyond the possibility of not being able to start a distributed
transaction?
Replies below...
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://www.msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"Sean" <spowell71@.hotmail.com> wrote in message
news:1126723915.556530.64310@.g44g2000cwa.googlegro ups.com...
> In Rodney R. Fournier's blog it states:
> "If MSDTC is used with SQL, you should ensure it is in it's own group,
> on it's own disk resource. Lastly, Microsoft recommends to run the
> MSDTC & SQL group on the same node, and that the MSDTC is brought
> online first when the groups are moved."
> ("http://msmvps.com/clustering/archive/2004/10/25/16672.aspx").
> I have a couple of questions about this but let first list my
> assumptions.
> Best:
> For high performance, installing an MSDTC Resource to it in its own
> group. This group should have its own disk and IP resources.
>
Correct.
> Good:
> The next best method is to put the MSDTC resource in the cluster or
> quorum group.
>
Correct.
> Worse:
> Put the MSDTC resource in SQL Group.
>
Yuck! Correct.
> What are the reasons for running the MSDTC & SQL group on the same
> node, beyond these potential Win2003 cluster issues listed in:
> http://support.microsoft.com/default...b;en-us;889706
> http://support.microsoft.com/kb/821751/
> ?
>
Correct, pre SP1 for Windows Server 2003 MSDTC has issues! Lots of issues.
The Q is correct.
> If DTC is in the quorum group is it still best to put SQL and quorum
> group on the same node?
>
If you can, yes!
> Our system admin is religious about putting the SQL group on a
> different node from the Quorum & DTC groups. Is there a best practice
> here or a KB I can point to?
>
http://support.microsoft.com/default...b;en-us;889706
Remember, SQL does not require MSDTC. You only need it if you application
will use it.
> Are there any implications not having SQL dependent of MSDTC starting
> first (MSDTC is brought online first when the groups are moved)
> beyond the possibility of not being able to start a distributed
> transaction?
>
Isn't that enough?
|||What are the arguments against these types of statements?
Having Quorum & DTC groups on a different node than the SQL group
a) has a load balancing effect, i.e. the SQL group will have the full
physical resources of the node.
b) enables the failover of SQL group happen be quicker cause Quorum &
DTC groups are already on the failover node.
|||http://support.microsoft.com/default...b;en-us;889706 pretty much
covers it.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering Website
http://www.msmvps.com/clustering - Blog
http://www.clusterhelp.com - Cluster Training
"Sean" <spowell71@.hotmail.com> wrote in message
news:1126789871.591787.32620@.g44g2000cwa.googlegro ups.com...
> What are the arguments against these types of statements?
> Having Quorum & DTC groups on a different node than the SQL group
> a) has a load balancing effect, i.e. the SQL group will have the full
> physical resources of the node.
> b) enables the failover of SQL group happen be quicker cause Quorum &
> DTC groups are already on the failover node.
>
|||I had this problem, MSTDC not available so I put MSDTC in the same group as
SQLServer AND made SQLServer dependant on MSDTC Works fine now.
Nik Marshall-Blank MCSD/MCDBA
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:%23sx340fuFHA.3188@.TK2MSFTNGP14.phx.gbl...
> http://support.microsoft.com/default...b;en-us;889706 pretty much
> covers it.
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering Website
> http://www.msmvps.com/clustering - Blog
> http://www.clusterhelp.com - Cluster Training
>
> "Sean" <spowell71@.hotmail.com> wrote in message
> news:1126789871.591787.32620@.g44g2000cwa.googlegro ups.com...
>
Showing posts with label practice. Show all posts
Showing posts with label practice. Show all posts
Friday, March 9, 2012
Wednesday, March 7, 2012
is it a good practice to do integrity check before perform backup?
On maintenance plan, there is a check box on integrity tab page, is it a good
way to do that first? if there is error found, will the backup stop? thanks
Hi
It depends on what error it will find. Actually, you can checkbox but make
sure that you uncheck Attempt to repair... option.
"tulip" <tulip@.discussions.microsoft.com> wrote in message
news:06E26A1F-4F08-4D81-9502-07E9DB0F5041@.microsoft.com...
> On maintenance plan, there is a check box on integrity tab page, is it a
> good
> way to do that first? if there is error found, will the backup stop?
> thanks
|||My approach is that if you have plenty of maintenance window for doing the
integraity checks (i.e. DBCC checks), do them by all means. And if you are
running into integrity issues, you may want to perform the DBCC checks more
regularly. It becomes a more difficult question when you are dealing with a
large amount of data and when you are struggling to contain your maintenance
jobs inside a limited maintenance window. In that case, you may have to forgo
the DBCC checks or do them very selectively.
Linchi
"tulip" wrote:
> On maintenance plan, there is a check box on integrity tab page, is it a good
> way to do that first? if there is error found, will the backup stop? thanks
way to do that first? if there is error found, will the backup stop? thanks
Hi
It depends on what error it will find. Actually, you can checkbox but make
sure that you uncheck Attempt to repair... option.
"tulip" <tulip@.discussions.microsoft.com> wrote in message
news:06E26A1F-4F08-4D81-9502-07E9DB0F5041@.microsoft.com...
> On maintenance plan, there is a check box on integrity tab page, is it a
> good
> way to do that first? if there is error found, will the backup stop?
> thanks
|||My approach is that if you have plenty of maintenance window for doing the
integraity checks (i.e. DBCC checks), do them by all means. And if you are
running into integrity issues, you may want to perform the DBCC checks more
regularly. It becomes a more difficult question when you are dealing with a
large amount of data and when you are struggling to contain your maintenance
jobs inside a limited maintenance window. In that case, you may have to forgo
the DBCC checks or do them very selectively.
Linchi
"tulip" wrote:
> On maintenance plan, there is a check box on integrity tab page, is it a good
> way to do that first? if there is error found, will the backup stop? thanks
is it a good practice to do integrity check before perform backup?
On maintenance plan, there is a check box on integrity tab page, is it a good
way to do that first? if there is error found, will the backup stop? thanksHi
It depends on what error it will find. Actually, you can checkbox but make
sure that you uncheck Attempt to repair... option.
"tulip" <tulip@.discussions.microsoft.com> wrote in message
news:06E26A1F-4F08-4D81-9502-07E9DB0F5041@.microsoft.com...
> On maintenance plan, there is a check box on integrity tab page, is it a
> good
> way to do that first? if there is error found, will the backup stop?
> thanks|||My approach is that if you have plenty of maintenance window for doing the
integraity checks (i.e. DBCC checks), do them by all means. And if you are
running into integrity issues, you may want to perform the DBCC checks more
regularly. It becomes a more difficult question when you are dealing with a
large amount of data and when you are struggling to contain your maintenance
jobs inside a limited maintenance window. In that case, you may have to forgo
the DBCC checks or do them very selectively.
Linchi
"tulip" wrote:
> On maintenance plan, there is a check box on integrity tab page, is it a good
> way to do that first? if there is error found, will the backup stop? thanks
way to do that first? if there is error found, will the backup stop? thanksHi
It depends on what error it will find. Actually, you can checkbox but make
sure that you uncheck Attempt to repair... option.
"tulip" <tulip@.discussions.microsoft.com> wrote in message
news:06E26A1F-4F08-4D81-9502-07E9DB0F5041@.microsoft.com...
> On maintenance plan, there is a check box on integrity tab page, is it a
> good
> way to do that first? if there is error found, will the backup stop?
> thanks|||My approach is that if you have plenty of maintenance window for doing the
integraity checks (i.e. DBCC checks), do them by all means. And if you are
running into integrity issues, you may want to perform the DBCC checks more
regularly. It becomes a more difficult question when you are dealing with a
large amount of data and when you are struggling to contain your maintenance
jobs inside a limited maintenance window. In that case, you may have to forgo
the DBCC checks or do them very selectively.
Linchi
"tulip" wrote:
> On maintenance plan, there is a check box on integrity tab page, is it a good
> way to do that first? if there is error found, will the backup stop? thanks
Is is a good practice to use autoincrement in access
Hi just new here in the forum i have a question
is it a good practice to use autoincrement fields in an access database for master and detail? and for unique record identification. if not please tell why..
im currently using access and with autoincrement for unique record identification and for master and detail.. then there is a possibilities to transfer from access to mssql. will i encounter problems when migrating since i am using an autoincrement fields
in mssql is there an autoincrement fields like access?
dont be harsh ok.. just anoob trying to learn. :eek:The equivalent of autoincrement in SQL server is the IDENTITY property.
It is possible to port existing ID values into a SQL Server table with an identity property set, but to do so you will need to temporarily turn off the identity property, and then reset it when the data load is complete.
The upsizing wizard for Access -> SQL Server may handle this for you.|||The equivalent of autoincrement in SQL server is the IDENTITY property.
It is possible to port existing ID values into a SQL Server table with an identity property set, but to do so you will need to temporarily turn off the identity property, and then reset it when the data load is complete.
The upsizing wizard for Access -> SQL Server may handle this for you.
thanks for the fast reply blindman.. but, is it a good practice to use autoincrement in a master detail ?? my master table will have an autoincrement fields, my detail will have a longint field to accomodate for the master's autoincrement field.
i will be using this method on a library program ..|||What is going to be the primary key of the detail table? A natural composite key including the master ID, or are you going to create a new surrogate key (possibly autoincremented)?|||What is going to be the primary key of the detail table? A natural composite key including the master ID, or are you going to create a new surrogate key (possibly autoincremented)?
the primary key of the detail table will be another autoincrement fields but with no relation whatsoever on the master field.|||An surrogate key on a detail table is not required, though it may facilitate SQL coding and programming the interface. You can leave it off and still have functional database design using the natural key of the detail table. Note that the natural key of the detail table is usually a composite of the surrogate key of the master table and some detail element that is unique within each master record.
One other thing; you keep asking whether it is good practice to use an autoincrementing column for a key. In SQL Server you basically have two choices for surrogate keys, and those are auto-incrementing identity values and GUIDs (Globally Unique Identifiers). Stick to using either of these. By definition surrogate keys should bear no relation to the underlying data, so you shouldn't be spending time mucking arround with methods of generating them. Hence, auto-generated surrogate keys such as Idenity columns and GUIDs are preferred.|||An surrogate key on a detail table is not required, though it may facilitate SQL coding and programming the interface. You can leave it off and still have functional database design using the natural key of the detail table. Note that the natural key of the detail table is usually a composite of the surrogate key of the master table and some detail element that is unique within each master record.
One other thing; you keep asking whether it is good practice to use an autoincrementing column for a key. In SQL Server you basically have two choices for surrogate keys, and those are auto-incrementing identity values and GUIDs (Globally Unique Identifiers). Stick to using either of these. By definition surrogate keys should bear no relation to the underlying data, so you shouldn't be spending time mucking arround with methods of generating them. Hence, auto-generated surrogate keys such as Idenity columns and GUIDs are preferred.
thanks.. i thought that i am at a lost.. thanks for clarifying those things. at least now i can sleep well coz i wont change the autoincrement fields into my own generated nos.. thanks again blindman. your a great help..
is it a good practice to use autoincrement fields in an access database for master and detail? and for unique record identification. if not please tell why..
im currently using access and with autoincrement for unique record identification and for master and detail.. then there is a possibilities to transfer from access to mssql. will i encounter problems when migrating since i am using an autoincrement fields
in mssql is there an autoincrement fields like access?
dont be harsh ok.. just anoob trying to learn. :eek:The equivalent of autoincrement in SQL server is the IDENTITY property.
It is possible to port existing ID values into a SQL Server table with an identity property set, but to do so you will need to temporarily turn off the identity property, and then reset it when the data load is complete.
The upsizing wizard for Access -> SQL Server may handle this for you.|||The equivalent of autoincrement in SQL server is the IDENTITY property.
It is possible to port existing ID values into a SQL Server table with an identity property set, but to do so you will need to temporarily turn off the identity property, and then reset it when the data load is complete.
The upsizing wizard for Access -> SQL Server may handle this for you.
thanks for the fast reply blindman.. but, is it a good practice to use autoincrement in a master detail ?? my master table will have an autoincrement fields, my detail will have a longint field to accomodate for the master's autoincrement field.
i will be using this method on a library program ..|||What is going to be the primary key of the detail table? A natural composite key including the master ID, or are you going to create a new surrogate key (possibly autoincremented)?|||What is going to be the primary key of the detail table? A natural composite key including the master ID, or are you going to create a new surrogate key (possibly autoincremented)?
the primary key of the detail table will be another autoincrement fields but with no relation whatsoever on the master field.|||An surrogate key on a detail table is not required, though it may facilitate SQL coding and programming the interface. You can leave it off and still have functional database design using the natural key of the detail table. Note that the natural key of the detail table is usually a composite of the surrogate key of the master table and some detail element that is unique within each master record.
One other thing; you keep asking whether it is good practice to use an autoincrementing column for a key. In SQL Server you basically have two choices for surrogate keys, and those are auto-incrementing identity values and GUIDs (Globally Unique Identifiers). Stick to using either of these. By definition surrogate keys should bear no relation to the underlying data, so you shouldn't be spending time mucking arround with methods of generating them. Hence, auto-generated surrogate keys such as Idenity columns and GUIDs are preferred.|||An surrogate key on a detail table is not required, though it may facilitate SQL coding and programming the interface. You can leave it off and still have functional database design using the natural key of the detail table. Note that the natural key of the detail table is usually a composite of the surrogate key of the master table and some detail element that is unique within each master record.
One other thing; you keep asking whether it is good practice to use an autoincrementing column for a key. In SQL Server you basically have two choices for surrogate keys, and those are auto-incrementing identity values and GUIDs (Globally Unique Identifiers). Stick to using either of these. By definition surrogate keys should bear no relation to the underlying data, so you shouldn't be spending time mucking arround with methods of generating them. Hence, auto-generated surrogate keys such as Idenity columns and GUIDs are preferred.
thanks.. i thought that i am at a lost.. thanks for clarifying those things. at least now i can sleep well coz i wont change the autoincrement fields into my own generated nos.. thanks again blindman. your a great help..
Friday, February 24, 2012
Is generating replica at night a best practice and what is the cos
Generating 5 fresh different publications (each hold data for one distinct
region) over night, so client can download their selection replica.
Now what is the cost for such thing?
What is the over head per 1MB of data in terms of system tables and in terms
of download times on a typical intranet/internet.
Personal experience will be excellent input.
How fast is the connection? How fast are the boxes? What type of
Replication? Lots of variables. Test, test, test.
"Lottoman2000 NEWBE" <Lottoman2000NEWBE@.discussions.microsoft.com> wrote in
message news:A252FA00-25CE-413D-BDE0-7830BF3A7DD5@.microsoft.com...
> Generating 5 fresh different publications (each hold data for one distinct
> region) over night, so client can download their selection replica.
> Now what is the cost for such thing?
> What is the over head per 1MB of data in terms of system tables and in
terms
> of download times on a typical intranet/internet.
> Personal experience will be excellent input.
>
region) over night, so client can download their selection replica.
Now what is the cost for such thing?
What is the over head per 1MB of data in terms of system tables and in terms
of download times on a typical intranet/internet.
Personal experience will be excellent input.
How fast is the connection? How fast are the boxes? What type of
Replication? Lots of variables. Test, test, test.
"Lottoman2000 NEWBE" <Lottoman2000NEWBE@.discussions.microsoft.com> wrote in
message news:A252FA00-25CE-413D-BDE0-7830BF3A7DD5@.microsoft.com...
> Generating 5 fresh different publications (each hold data for one distinct
> region) over night, so client can download their selection replica.
> Now what is the cost for such thing?
> What is the over head per 1MB of data in terms of system tables and in
terms
> of download times on a typical intranet/internet.
> Personal experience will be excellent input.
>
Subscribe to:
Posts (Atom)