Hi,
We have an active-active cluster where SQL server instances are created on
node 1 and Analysis services are installed on node 2. We need to upgrade the
H/W for both nodes? Can I do a rolling upgrade i.e. replacing one node at
time in above configuration?
If not then what are my options & if yes then can you give me some links to
documentation on how to do this?
Thanks,
Sushil
Sushil
Yes you can. I have done exactly this type of project. The more you plan
and test, the better your chance of success. I don't have any publicly
available documentation on a rolling upgrade.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"Sushil" <Sushil@.discussions.microsoft.com> wrote in message
news:EA4D680F-4548-4536-8BD2-88ECE1004CF8@.microsoft.com...
> Hi,
> We have an active-active cluster where SQL server instances are created on
> node 1 and Analysis services are installed on node 2. We need to upgrade
> the
> H/W for both nodes? Can I do a rolling upgrade i.e. replacing one node at
> time in above configuration?
> If not then what are my options & if yes then can you give me some links
> to
> documentation on how to do this?
> Thanks,
> --
> Sushil
> --
> Sushil
|||Geoff,
Thanks for response. Can you please list if you had any challenges while
doing this?
I have been told that, When we replace node 2 by new node (Just with OS)
and install SQL server on this new node and try to cluster this new node
with OLD node 1 then that clustering will not work.
Can you point me any Microsoft documentation or books where I can get some
clues?
Thanks,
Sushil
Sushil
"Geoff N. Hiten" wrote:
> Yes you can. I have done exactly this type of project. The more you plan
> and test, the better your chance of success. I don't have any publicly
> available documentation on a rolling upgrade.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
>
> "Sushil" <Sushil@.discussions.microsoft.com> wrote in message
> news:EA4D680F-4548-4536-8BD2-88ECE1004CF8@.microsoft.com...
>
>
|||Here is the quick version.
Starting with SQL 2000 2-node (4-Proc DELL) connected to DELL | EMC CX-600.
Windows 2000 OS. 2 SQL Instances.
Ended with SQL 2000 4-node (8-proc Unisys) still connected to the same SAN
(expanded the storage capacity, though). Windows 2003 OS. 3 SQL Instances.
Started with in-place upgrade of OS (one node at a time). Obtaining the
correct drivers for the HBAs was a challenge. This was Q4 '03, so that
shouldn't be a problem, but it is a critical step.
Added the two new nodes. When stable, removed original two nodes. Finally
added the remaining new nodes.
Note that if you have a SCSI based cluster, you cannot have more than two
nodes even with Windows 2003.
Again, planning and testing was essential. Each step had numerous tests and
bailout plans to avoid downtime.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"Sushil" <Sushil@.discussions.microsoft.com> wrote in message
news:97D3A989-1C04-450C-B3C4-20F674299E09@.microsoft.com...[vbcol=seagreen]
> Geoff,
> Thanks for response. Can you please list if you had any challenges while
> doing this?
> I have been told that, When we replace node 2 by new node (Just with OS)
> and install SQL server on this new node and try to cluster this new node
> with OLD node 1 then that clustering will not work.
> Can you point me any Microsoft documentation or books where I can get some
> clues?
> Thanks,
> Sushil
>
> --
> Sushil
>
> "Geoff N. Hiten" wrote:
|||I have a thought rolling around in my brain about limitations of the number
of nodes being different between WIn2K and Win2K3...but I can't remember the
specifics. Geoff, can you recall or am I off-base (anymore than usual?)
Kevin Hill
3NF Consulting
www.3nf-inc.com/NewsGroups.htm
www.expertsrt.com - not your average tech Q&A site
"Geoff N. Hiten" <SQLCraftsman@.gmail.com> wrote in message
news:OF3pKh0MGHA.4052@.TK2MSFTNGP15.phx.gbl...
> Here is the quick version.
> Starting with SQL 2000 2-node (4-Proc DELL) connected to DELL | EMC
> CX-600. Windows 2000 OS. 2 SQL Instances.
> Ended with SQL 2000 4-node (8-proc Unisys) still connected to the same SAN
> (expanded the storage capacity, though). Windows 2003 OS. 3 SQL
> Instances.
> Started with in-place upgrade of OS (one node at a time). Obtaining the
> correct drivers for the HBAs was a challenge. This was Q4 '03, so that
> shouldn't be a problem, but it is a critical step.
> Added the two new nodes. When stable, removed original two nodes.
> Finally added the remaining new nodes.
> Note that if you have a SCSI based cluster, you cannot have more than two
> nodes even with Windows 2003.
> Again, planning and testing was essential. Each step had numerous tests
> and bailout plans to avoid downtime.
> --
> Geoff N. Hiten
> Senior Database Administrator
> Microsoft SQL Server MVP
>
> "Sushil" <Sushil@.discussions.microsoft.com> wrote in message
> news:97D3A989-1C04-450C-B3C4-20F674299E09@.microsoft.com...
>
|||Windows 2000 Enterprise Edition supports 2 nodes. Windows 2003 supports up
to 8 nodes, but SQL 2000 Enterprise Edition only supports a maximum of 4
nodes so that is the limit. That is why I had to do an in-place OS upgrade
as a beginning step for this process. I normally don't like in-place
upgrades, but this was only for a short time during a transition.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"Kevin3NF" <Kevin@.DontNeedNoSpam3NF-inc.com> wrote in message
news:eP$dgeWNGHA.1192@.TK2MSFTNGP11.phx.gbl...
>I have a thought rolling around in my brain about limitations of the number
>of nodes being different between WIn2K and Win2K3...but I can't remember
>the specifics. Geoff, can you recall or am I off-base (anymore than
>usual?)
> --
> Kevin Hill
> 3NF Consulting
> www.3nf-inc.com/NewsGroups.htm
> www.expertsrt.com - not your average tech Q&A site
>
> "Geoff N. Hiten" <SQLCraftsman@.gmail.com> wrote in message
> news:OF3pKh0MGHA.4052@.TK2MSFTNGP15.phx.gbl...
>
Showing posts with label cluster. Show all posts
Showing posts with label cluster. Show all posts
Sunday, March 25, 2012
Thursday, March 22, 2012
Can I create another one virtual SQL Server at the same cluster
Hi,
I already have one virtual SQL server in the cluster, it's owned by only one
node.
Now I added another node into the cluster, and we don't want this node to be
a passive node, and don't want go to multi-instance .
So, the question is if
I can create another one virtual SQL server on the new node, give it new
SQL IP,
SQL Network Name, SQL Server...
Please advise.
Thanks you so much.
Peter
You sure can, you can have 16 instances per cluster when running on Windows
Server 2003.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> Hi,
> I already have one virtual SQL server in the cluster, it's owned by only
> one
> node.
> Now I added another node into the cluster, and we don't want this node to
> be
> a passive node, and don't want go to multi-instance .
> So, the question is if
> I can create another one virtual SQL server on the new node, give it new
> SQL IP,
> SQL Network Name, SQL Server...
> Please advise.
> Thanks you so much.
> Peter
>
>
|||Thank you so much, Rod !!
So, I can do the same steps on the new node to create the new virtual SQL as
what I did on the older node before.
Thanks,
Peter
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> Hi,
> I already have one virtual SQL server in the cluster, it's owned by only
one
> node.
> Now I added another node into the cluster, and we don't want this node to
be
> a passive node, and don't want go to multi-instance .
> So, the question is if
> I can create another one virtual SQL server on the new node, give it new
> SQL IP,
> SQL Network Name, SQL Server...
> Please advise.
> Thanks you so much.
> Peter
>
>
|||Yes, only thing to watch out is that you can only have 1 default instance,
the rest are named. Or you can have all named instances, either way 16 is
the limit.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e%23HWWUJNFHA.1096@.tk2msftngp13.phx.gbl...
> Thank you so much, Rod !!
> So, I can do the same steps on the new node to create the new virtual SQL
> as
> what I did on the older node before.
> Thanks,
> Peter
> "Koarla SUN" <koarla@.hotmail.com> wrote in message
> news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> one
> be
>
|||That's mean I had to uncheck the default box and enter the instance name in
the [Instance Name] step.
So, new virtual SQL name will be "preview default instance name\instance
name"
And I can still use the default TCP port for the new virtual SQL instance.
Thank you, Rod.
Peter
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:u1KyWeJNFHA.3708@.TK2MSFTNGP14.phx.gbl...[vbcol=seagreen]
> Yes, only thing to watch out is that you can only have 1 default instance,
> the rest are named. Or you can have all named instances, either way 16 is
> the limit.
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Koarla SUN" <koarla@.hotmail.com> wrote in message
> news:e%23HWWUJNFHA.1096@.tk2msftngp13.phx.gbl...
SQL[vbcol=seagreen]
only[vbcol=seagreen]
to[vbcol=seagreen]
new
>
|||Correct. Remember its a new name and IP so the ports will rename the same.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Peter Sun" <koarla@.hotmail.com> wrote in message
news:O9caotJNFHA.580@.TK2MSFTNGP15.phx.gbl...
> That's mean I had to uncheck the default box and enter the instance name
> in
> the [Instance Name] step.
> So, new virtual SQL name will be "preview default instance name\instance
> name"
> And I can still use the default TCP port for the new virtual SQL instance.
> Thank you, Rod.
> Peter
>
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:u1KyWeJNFHA.3708@.TK2MSFTNGP14.phx.gbl...
> SQL
> only
> to
> new
>
|||I was confused with the multi-instance, I was think I can go ahead to create
a another one virtual SQL server, just put these virtual SQL servers under
the same cluster.
So, I was thinking I enter a different viirtual SQL server name at the first
step, and still check default box in the [instance name} step. Then we will
have kind of virtual SQL server naming, Node1 , Node 2, ...Node 16
As I assign a new IP with the new virtual SQL server, I should be able to
use the default SQL tcp port : 1433.
I appreciate yoou got me clear.
Thank you so much.
Peter
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:ulLDR2JNFHA.688@.TK2MSFTNGP10.phx.gbl...
> Correct. Remember its a new name and IP so the ports will rename the
same.[vbcol=seagreen]
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Peter Sun" <koarla@.hotmail.com> wrote in message
> news:O9caotJNFHA.580@.TK2MSFTNGP15.phx.gbl...
name[vbcol=seagreen]
instance.[vbcol=seagreen]
is[vbcol=seagreen]
node[vbcol=seagreen]
it
>
I already have one virtual SQL server in the cluster, it's owned by only one
node.
Now I added another node into the cluster, and we don't want this node to be
a passive node, and don't want go to multi-instance .
So, the question is if
I can create another one virtual SQL server on the new node, give it new
SQL IP,
SQL Network Name, SQL Server...
Please advise.
Thanks you so much.
Peter
You sure can, you can have 16 instances per cluster when running on Windows
Server 2003.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> Hi,
> I already have one virtual SQL server in the cluster, it's owned by only
> one
> node.
> Now I added another node into the cluster, and we don't want this node to
> be
> a passive node, and don't want go to multi-instance .
> So, the question is if
> I can create another one virtual SQL server on the new node, give it new
> SQL IP,
> SQL Network Name, SQL Server...
> Please advise.
> Thanks you so much.
> Peter
>
>
|||Thank you so much, Rod !!
So, I can do the same steps on the new node to create the new virtual SQL as
what I did on the older node before.
Thanks,
Peter
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> Hi,
> I already have one virtual SQL server in the cluster, it's owned by only
one
> node.
> Now I added another node into the cluster, and we don't want this node to
be
> a passive node, and don't want go to multi-instance .
> So, the question is if
> I can create another one virtual SQL server on the new node, give it new
> SQL IP,
> SQL Network Name, SQL Server...
> Please advise.
> Thanks you so much.
> Peter
>
>
|||Yes, only thing to watch out is that you can only have 1 default instance,
the rest are named. Or you can have all named instances, either way 16 is
the limit.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Koarla SUN" <koarla@.hotmail.com> wrote in message
news:e%23HWWUJNFHA.1096@.tk2msftngp13.phx.gbl...
> Thank you so much, Rod !!
> So, I can do the same steps on the new node to create the new virtual SQL
> as
> what I did on the older node before.
> Thanks,
> Peter
> "Koarla SUN" <koarla@.hotmail.com> wrote in message
> news:e7jTTLJNFHA.3900@.TK2MSFTNGP10.phx.gbl...
> one
> be
>
|||That's mean I had to uncheck the default box and enter the instance name in
the [Instance Name] step.
So, new virtual SQL name will be "preview default instance name\instance
name"
And I can still use the default TCP port for the new virtual SQL instance.
Thank you, Rod.
Peter
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:u1KyWeJNFHA.3708@.TK2MSFTNGP14.phx.gbl...[vbcol=seagreen]
> Yes, only thing to watch out is that you can only have 1 default instance,
> the rest are named. Or you can have all named instances, either way 16 is
> the limit.
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Koarla SUN" <koarla@.hotmail.com> wrote in message
> news:e%23HWWUJNFHA.1096@.tk2msftngp13.phx.gbl...
SQL[vbcol=seagreen]
only[vbcol=seagreen]
to[vbcol=seagreen]
new
>
|||Correct. Remember its a new name and IP so the ports will rename the same.
Cheers,
Rod
MVP - Windows Server - Clustering
http://www.nw-america.com - Clustering
http://msmvps.com/clustering - Blog
"Peter Sun" <koarla@.hotmail.com> wrote in message
news:O9caotJNFHA.580@.TK2MSFTNGP15.phx.gbl...
> That's mean I had to uncheck the default box and enter the instance name
> in
> the [Instance Name] step.
> So, new virtual SQL name will be "preview default instance name\instance
> name"
> And I can still use the default TCP port for the new virtual SQL instance.
> Thank you, Rod.
> Peter
>
> "Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
> message news:u1KyWeJNFHA.3708@.TK2MSFTNGP14.phx.gbl...
> SQL
> only
> to
> new
>
|||I was confused with the multi-instance, I was think I can go ahead to create
a another one virtual SQL server, just put these virtual SQL servers under
the same cluster.
So, I was thinking I enter a different viirtual SQL server name at the first
step, and still check default box in the [instance name} step. Then we will
have kind of virtual SQL server naming, Node1 , Node 2, ...Node 16
As I assign a new IP with the new virtual SQL server, I should be able to
use the default SQL tcp port : 1433.
I appreciate yoou got me clear.
Thank you so much.
Peter
"Rodney R. Fournier [MVP]" <rod@.die.spam.die.nw-america.com> wrote in
message news:ulLDR2JNFHA.688@.TK2MSFTNGP10.phx.gbl...
> Correct. Remember its a new name and IP so the ports will rename the
same.[vbcol=seagreen]
> Cheers,
> Rod
> MVP - Windows Server - Clustering
> http://www.nw-america.com - Clustering
> http://msmvps.com/clustering - Blog
> "Peter Sun" <koarla@.hotmail.com> wrote in message
> news:O9caotJNFHA.580@.TK2MSFTNGP15.phx.gbl...
name[vbcol=seagreen]
instance.[vbcol=seagreen]
is[vbcol=seagreen]
node[vbcol=seagreen]
it
>
Tuesday, March 20, 2012
Can I combine Log Shipping and Clustering?
SQL Server 2005 SP2
Certainly. They're two completely unrelated features.
--
Greg Moore
SQL Server DBA Consulting Remote and Onsite available!
Email: sql (at) greenms.com http://www.greenms.com/sqlserver.html
Can I combine Log Shipping and Clustering? That is, can I
ship logs from the primary node in my cluster to an offsite
warm standby SQL server box?
Any help appreciated.
TIA
aj"aj" <ronald@.mcdonalds.comwrote in message
news:13k3vep1gmodlec@.news.supernews.com...
Quote:
Originally Posted by
SQL Server 2005 SP2
>
Can I combine Log Shipping and Clustering? That is, can I
ship logs from the primary node in my cluster to an offsite
warm standby SQL server box?
>
Certainly. They're two completely unrelated features.
And it's fairly common to in fact do what you propose.
Quote:
Originally Posted by
Any help appreciated.
>
TIA
>
aj
--
Greg Moore
SQL Server DBA Consulting Remote and Onsite available!
Email: sql (at) greenms.com http://www.greenms.com/sqlserver.html
Can I change the SQL 2000 Cluster database from case-sensitive to case-insensitive in the
Hi All,
I'd like to know if I can change the SQL 2000 Cluster database configuration
from case-sensitive to case-insensitive in the running box. If yes, how can
I do that?
TIACould you please send more detail and/or one example of what you are trying
to accomplish?
Ben Nevarez, MCDBA, OCP
Database Administrator
"M C" wrote:
> Hi All,
> I'd like to know if I can change the SQL 2000 Cluster database configuration
> from case-sensitive to case-insensitive in the running box. If yes, how can
> I do that?
> TIA
>
>|||Hi ALL,
When I first installed SQL 2000 Cluster, I choose the case-sensitive when
the setup prompt for collation. I'm now want to change the collation to
case-insensitive. Can I do that?
TIA.
"Ben Nevarez" <bnevarez@.sjm.com> wrote in message
news:404BF5E3-E8E0-420A-9193-6E41C93B2DE3@.microsoft.com...
> Could you please send more detail and/or one example of what you are
> trying
> to accomplish?
> Ben Nevarez, MCDBA, OCP
> Database Administrator
>
> "M C" wrote:
>> Hi All,
>> I'd like to know if I can change the SQL 2000 Cluster database
>> configuration
>> from case-sensitive to case-insensitive in the running box. If yes, how
>> can
>> I do that?
>> TIA
>>|||"M C" <abc@.def.com> wrote in message
news:umNMiyomGHA.3732@.TK2MSFTNGP05.phx.gbl...
> Hi ALL,
> When I first installed SQL 2000 Cluster, I choose the case-sensitive when
> the setup prompt for collation. I'm now want to change the collation to
> case-insensitive. Can I do that?
Not on a server-wide basis w/o doing a complete re-install.
On a per database basis you can, but it can get messy.
> TIA.
> "Ben Nevarez" <bnevarez@.sjm.com> wrote in message
> news:404BF5E3-E8E0-420A-9193-6E41C93B2DE3@.microsoft.com...
> >
> > Could you please send more detail and/or one example of what you are
> > trying
> > to accomplish?
> >
> > Ben Nevarez, MCDBA, OCP
> > Database Administrator
> >
> >
> > "M C" wrote:
> >
> >> Hi All,
> >>
> >> I'd like to know if I can change the SQL 2000 Cluster database
> >> configuration
> >> from case-sensitive to case-insensitive in the running box. If yes, how
> >> can
> >> I do that?
> >>
> >> TIA
> >>
> >>
> >>
>sql
I'd like to know if I can change the SQL 2000 Cluster database configuration
from case-sensitive to case-insensitive in the running box. If yes, how can
I do that?
TIACould you please send more detail and/or one example of what you are trying
to accomplish?
Ben Nevarez, MCDBA, OCP
Database Administrator
"M C" wrote:
> Hi All,
> I'd like to know if I can change the SQL 2000 Cluster database configuration
> from case-sensitive to case-insensitive in the running box. If yes, how can
> I do that?
> TIA
>
>|||Hi ALL,
When I first installed SQL 2000 Cluster, I choose the case-sensitive when
the setup prompt for collation. I'm now want to change the collation to
case-insensitive. Can I do that?
TIA.
"Ben Nevarez" <bnevarez@.sjm.com> wrote in message
news:404BF5E3-E8E0-420A-9193-6E41C93B2DE3@.microsoft.com...
> Could you please send more detail and/or one example of what you are
> trying
> to accomplish?
> Ben Nevarez, MCDBA, OCP
> Database Administrator
>
> "M C" wrote:
>> Hi All,
>> I'd like to know if I can change the SQL 2000 Cluster database
>> configuration
>> from case-sensitive to case-insensitive in the running box. If yes, how
>> can
>> I do that?
>> TIA
>>|||"M C" <abc@.def.com> wrote in message
news:umNMiyomGHA.3732@.TK2MSFTNGP05.phx.gbl...
> Hi ALL,
> When I first installed SQL 2000 Cluster, I choose the case-sensitive when
> the setup prompt for collation. I'm now want to change the collation to
> case-insensitive. Can I do that?
Not on a server-wide basis w/o doing a complete re-install.
On a per database basis you can, but it can get messy.
> TIA.
> "Ben Nevarez" <bnevarez@.sjm.com> wrote in message
> news:404BF5E3-E8E0-420A-9193-6E41C93B2DE3@.microsoft.com...
> >
> > Could you please send more detail and/or one example of what you are
> > trying
> > to accomplish?
> >
> > Ben Nevarez, MCDBA, OCP
> > Database Administrator
> >
> >
> > "M C" wrote:
> >
> >> Hi All,
> >>
> >> I'd like to know if I can change the SQL 2000 Cluster database
> >> configuration
> >> from case-sensitive to case-insensitive in the running box. If yes, how
> >> can
> >> I do that?
> >>
> >> TIA
> >>
> >>
> >>
>sql
Labels:
box,
case-insensitive,
case-sensitive,
cluster,
configuration,
database,
microsoft,
mysql,
oracle,
running,
server,
sql
Saturday, February 25, 2012
Can anyone recommend the best resource for performance tuning a SQL cluster?
I have a couple clusters (each with eight processors total, and 16 GB RAM
total), but am not getting the performance desired. I "inherited" these, I
didn't build them. Can anyone recommend any resources on tuning the cluster
and tuning the SQL service for the best performance?
The Cluster itself has no bearing on performance. You would tune the DB's
just like any other. The only exception is possibly the amount of ram you
use if you have a multi-instance cluster. See here for a start:
http://www.microsoft.com/sql/techinf...perftuning.asp
Performance WP's
http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
http://www.sql-server-performance.co...ance_audit.asp
Hardware Performance CheckList
http://www.sql-server-performance.co...mance_tips.asp
SQL 2000 Performance tuning tips
http://www.support.microsoft.com/?id=q224587 Troubleshooting App
Performance
http://msdn.microsoft.com/library/de...rfmon_24u1.asp
Disk Monitoring
http://www.sql-server-performance.co...showcontig.asp Understanding
DBCC SHOWCONTIG
Andrew J. Kelly SQL MVP
"Adam Freden" <Adam.Freden@.cpsa-rbha.org> wrote in message
news:uUH3lnnmEHA.3356@.TK2MSFTNGP14.phx.gbl...
> I have a couple clusters (each with eight processors total, and 16 GB RAM
> total), but am not getting the performance desired. I "inherited" these,
I
> didn't build them. Can anyone recommend any resources on tuning the
cluster
> and tuning the SQL service for the best performance?
>
|||Just my two cents. Clusters are almost always built on advanced storage
technology (Except for SCSI-based clusters, which are evil). As such, the
performance of the storage device comes into play. If you find the system
is suffering from I/O limitations, you may need to dig into the SAN
performance and tuning to solve your problems. I would start with the links
Andrew sent and only go after SAN tuning once you have found and fixed all
your other performance issues.
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com
I support the Professional Association for SQL Server
www.sqlpass.org
"Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote in message
news:%23jH67KomEHA.3100@.TK2MSFTNGP10.phx.gbl...
> The Cluster itself has no bearing on performance. You would tune the DB's
> just like any other. The only exception is possibly the amount of ram you
> use if you have a multi-instance cluster. See here for a start:
>
> http://www.microsoft.com/sql/techinf...perftuning.asp
> Performance WP's
> http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
> http://www.sql-server-performance.co...ance_audit.asp
> Hardware Performance CheckList
> http://www.sql-server-performance.co...mance_tips.asp
> SQL 2000 Performance tuning tips
> http://www.support.microsoft.com/?id=q224587 Troubleshooting App
> Performance
>
http://msdn.microsoft.com/library/de...rfmon_24u1.asp
> Disk Monitoring
> http://www.sql-server-performance.co...showcontig.asp
Understanding[vbcol=seagreen]
> DBCC SHOWCONTIG
>
> --
> Andrew J. Kelly SQL MVP
>
> "Adam Freden" <Adam.Freden@.cpsa-rbha.org> wrote in message
> news:uUH3lnnmEHA.3356@.TK2MSFTNGP14.phx.gbl...
RAM[vbcol=seagreen]
these,
> I
> cluster
>
total), but am not getting the performance desired. I "inherited" these, I
didn't build them. Can anyone recommend any resources on tuning the cluster
and tuning the SQL service for the best performance?
The Cluster itself has no bearing on performance. You would tune the DB's
just like any other. The only exception is possibly the amount of ram you
use if you have a multi-instance cluster. See here for a start:
http://www.microsoft.com/sql/techinf...perftuning.asp
Performance WP's
http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
http://www.sql-server-performance.co...ance_audit.asp
Hardware Performance CheckList
http://www.sql-server-performance.co...mance_tips.asp
SQL 2000 Performance tuning tips
http://www.support.microsoft.com/?id=q224587 Troubleshooting App
Performance
http://msdn.microsoft.com/library/de...rfmon_24u1.asp
Disk Monitoring
http://www.sql-server-performance.co...showcontig.asp Understanding
DBCC SHOWCONTIG
Andrew J. Kelly SQL MVP
"Adam Freden" <Adam.Freden@.cpsa-rbha.org> wrote in message
news:uUH3lnnmEHA.3356@.TK2MSFTNGP14.phx.gbl...
> I have a couple clusters (each with eight processors total, and 16 GB RAM
> total), but am not getting the performance desired. I "inherited" these,
I
> didn't build them. Can anyone recommend any resources on tuning the
cluster
> and tuning the SQL service for the best performance?
>
|||Just my two cents. Clusters are almost always built on advanced storage
technology (Except for SCSI-based clusters, which are evil). As such, the
performance of the storage device comes into play. If you find the system
is suffering from I/O limitations, you may need to dig into the SAN
performance and tuning to solve your problems. I would start with the links
Andrew sent and only go after SAN tuning once you have found and fixed all
your other performance issues.
Geoff N. Hiten
Microsoft SQL Server MVP
Senior Database Administrator
Careerbuilder.com
I support the Professional Association for SQL Server
www.sqlpass.org
"Andrew J. Kelly" <sqlmvpnooospam@.shadhawk.com> wrote in message
news:%23jH67KomEHA.3100@.TK2MSFTNGP10.phx.gbl...
> The Cluster itself has no bearing on performance. You would tune the DB's
> just like any other. The only exception is possibly the amount of ram you
> use if you have a multi-instance cluster. See here for a start:
>
> http://www.microsoft.com/sql/techinf...perftuning.asp
> Performance WP's
> http://www.swynk.com/friends/vandenberg/perfmonitor.asp Perfmon counters
> http://www.sql-server-performance.co...ance_audit.asp
> Hardware Performance CheckList
> http://www.sql-server-performance.co...mance_tips.asp
> SQL 2000 Performance tuning tips
> http://www.support.microsoft.com/?id=q224587 Troubleshooting App
> Performance
>
http://msdn.microsoft.com/library/de...rfmon_24u1.asp
> Disk Monitoring
> http://www.sql-server-performance.co...showcontig.asp
Understanding[vbcol=seagreen]
> DBCC SHOWCONTIG
>
> --
> Andrew J. Kelly SQL MVP
>
> "Adam Freden" <Adam.Freden@.cpsa-rbha.org> wrote in message
> news:uUH3lnnmEHA.3356@.TK2MSFTNGP14.phx.gbl...
RAM[vbcol=seagreen]
these,
> I
> cluster
>
Sunday, February 19, 2012
Can a SQL "virtual server" contains more than one SQL instance ?
Hi, I am new to SQL clustering.
I have a two nodes windows 2003 enterprise server in a cluster with sql 2000
enterprise server installed. Following the SQL installation screen I have
create a virtual server ( for SQL ) and installed the first sql instance
using "default" as the name .So far no problem.
But when I tried to install another instance of SQL it seems like I could
only create a new Virtual server and then name another SQLinstance (
example: Virtual server is " ProdSQL" and instance is INST1) , so it becomes
prodsql\inst1.
My question ( confusion ) is : can I create another instance using the same
Virtual server "prodsql" and install another instance "inst2" so this will
become prodsql\inst2. Is this possible and how to do it ?( Or one cluster
group can have only one virtual server can only have one single instance of
SQL ? )
Any advice appreciated.
George Norman
From BOL ("usually" your best buddy):
Multiple Instances of SQL Server on a Failover Cluster
You can run only one instance of SQL Server on each virtual server of a SQL
Server failover cluster, although you can install up to 16 virtual servers
on a failover cluster. The instance can be either a default instance or a
named instance. The virtual server looks like a single computer to
applications connecting to that instance of SQL Server. When applications
connect to the virtual server, they use the same convention as when
connecting to any instance of SQL Server; they specify the virtual server
name of the cluster and the optional instance name (only needed for named
instances): virtualservername\instancename. For more information about
clustering, see Failover Clustering Architecture.
joe.
"Norman" <GeorgeNorman@.hotmail.com> wrote in message
news:emEeoMuYFHA.3572@.TK2MSFTNGP12.phx.gbl...
> Hi, I am new to SQL clustering.
> I have a two nodes windows 2003 enterprise server in a cluster with sql
> 2000 enterprise server installed. Following the SQL installation screen I
> have create a virtual server ( for SQL ) and installed the first sql
> instance using "default" as the name .So far no problem.
> But when I tried to install another instance of SQL it seems like I could
> only create a new Virtual server and then name another SQLinstance (
> example: Virtual server is " ProdSQL" and instance is INST1) , so it
> becomes prodsql\inst1.
> My question ( confusion ) is : can I create another instance using the
> same Virtual server "prodsql" and install another instance "inst2" so this
> will become prodsql\inst2. Is this possible and how to do it ?( Or one
> cluster group can have only one virtual server can only have one single
> instance of SQL ? )
> Any advice appreciated.
> George Norman
>
|||Thanks Joe, that clears things up.
So when Microsoft said Multiple instance it means "multiple Virtual Servers
each with a single SQL instance" .
So does it also means that the Active -Active ( say between two nodes ) is
just running one Virtual server on each of the nodes to spread the work load
( and of course one of the node will control the quorum )?
George
"Joe Yong" <jyongNOSPAM@.scalabilityexperts.com> wrote in message
news:eJ6tUeuYFHA.3356@.TK2MSFTNGP15.phx.gbl...
> From BOL ("usually" your best buddy):
> Multiple Instances of SQL Server on a Failover Cluster
> You can run only one instance of SQL Server on each virtual server of a
> SQL Server failover cluster, although you can install up to 16 virtual
> servers on a failover cluster. The instance can be either a default
> instance or a named instance. The virtual server looks like a single
> computer to applications connecting to that instance of SQL Server. When
> applications connect to the virtual server, they use the same convention
> as when connecting to any instance of SQL Server; they specify the virtual
> server name of the cluster and the optional instance name (only needed for
> named instances): virtualservername\instancename. For more information
> about clustering, see Failover Clustering Architecture.
>
>
> joe.
> "Norman" <GeorgeNorman@.hotmail.com> wrote in message
> news:emEeoMuYFHA.3572@.TK2MSFTNGP12.phx.gbl...
>
I have a two nodes windows 2003 enterprise server in a cluster with sql 2000
enterprise server installed. Following the SQL installation screen I have
create a virtual server ( for SQL ) and installed the first sql instance
using "default" as the name .So far no problem.
But when I tried to install another instance of SQL it seems like I could
only create a new Virtual server and then name another SQLinstance (
example: Virtual server is " ProdSQL" and instance is INST1) , so it becomes
prodsql\inst1.
My question ( confusion ) is : can I create another instance using the same
Virtual server "prodsql" and install another instance "inst2" so this will
become prodsql\inst2. Is this possible and how to do it ?( Or one cluster
group can have only one virtual server can only have one single instance of
SQL ? )
Any advice appreciated.
George Norman
From BOL ("usually" your best buddy):
Multiple Instances of SQL Server on a Failover Cluster
You can run only one instance of SQL Server on each virtual server of a SQL
Server failover cluster, although you can install up to 16 virtual servers
on a failover cluster. The instance can be either a default instance or a
named instance. The virtual server looks like a single computer to
applications connecting to that instance of SQL Server. When applications
connect to the virtual server, they use the same convention as when
connecting to any instance of SQL Server; they specify the virtual server
name of the cluster and the optional instance name (only needed for named
instances): virtualservername\instancename. For more information about
clustering, see Failover Clustering Architecture.
joe.
"Norman" <GeorgeNorman@.hotmail.com> wrote in message
news:emEeoMuYFHA.3572@.TK2MSFTNGP12.phx.gbl...
> Hi, I am new to SQL clustering.
> I have a two nodes windows 2003 enterprise server in a cluster with sql
> 2000 enterprise server installed. Following the SQL installation screen I
> have create a virtual server ( for SQL ) and installed the first sql
> instance using "default" as the name .So far no problem.
> But when I tried to install another instance of SQL it seems like I could
> only create a new Virtual server and then name another SQLinstance (
> example: Virtual server is " ProdSQL" and instance is INST1) , so it
> becomes prodsql\inst1.
> My question ( confusion ) is : can I create another instance using the
> same Virtual server "prodsql" and install another instance "inst2" so this
> will become prodsql\inst2. Is this possible and how to do it ?( Or one
> cluster group can have only one virtual server can only have one single
> instance of SQL ? )
> Any advice appreciated.
> George Norman
>
|||Thanks Joe, that clears things up.
So when Microsoft said Multiple instance it means "multiple Virtual Servers
each with a single SQL instance" .
So does it also means that the Active -Active ( say between two nodes ) is
just running one Virtual server on each of the nodes to spread the work load
( and of course one of the node will control the quorum )?
George
"Joe Yong" <jyongNOSPAM@.scalabilityexperts.com> wrote in message
news:eJ6tUeuYFHA.3356@.TK2MSFTNGP15.phx.gbl...
> From BOL ("usually" your best buddy):
> Multiple Instances of SQL Server on a Failover Cluster
> You can run only one instance of SQL Server on each virtual server of a
> SQL Server failover cluster, although you can install up to 16 virtual
> servers on a failover cluster. The instance can be either a default
> instance or a named instance. The virtual server looks like a single
> computer to applications connecting to that instance of SQL Server. When
> applications connect to the virtual server, they use the same convention
> as when connecting to any instance of SQL Server; they specify the virtual
> server name of the cluster and the optional instance name (only needed for
> named instances): virtualservername\instancename. For more information
> about clustering, see Failover Clustering Architecture.
>
>
> joe.
> "Norman" <GeorgeNorman@.hotmail.com> wrote in message
> news:emEeoMuYFHA.3572@.TK2MSFTNGP12.phx.gbl...
>
Tuesday, February 14, 2012
Can 2K5 Cluster can have 2 instances on each node?
Hi,
We have sql 2005 standard installed as cluster on win2k3 cluster with 8
GB memory on each node. We have one virtual sql server named instance on
first node. Can we create one more instance on the same node and
similarly 2 more instances on the second node. The cluster will be in
active\active mode. Our reason for considering this is that we will be
able to fail over one instance from the node 1 to node2 and node2 still
have the 2nd instance running. Is this possible?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
I think you have some misconceptions on how clustering works. The obsolete
naming of "Active/passive/Active..." is partially responsible for that.
Instances are installed to the cluster. You can have up to sixteen
instances on a cluster. SQL Server 2005 Standard Edition supports up to
two-node clusters. Nodes are the host computers for a cluster. An instance
can run on either node, but only on one node at a time. You cannot install
the same SQL instance to two nodes and have them both serviceing data
requests as if they were the same server. SQL Clustering is a failover
technology, not a scale-out technology. You can create two SQL instances on
a SQL 2005 two-node cluster and let one run normally on each node. During a
failure event one node can run both instances, with some performance
compromises of course.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"DeBe" <DeBe.25toh0@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25toh0@.no-mx.forums.yourdomain.com.au...
> Hi,
> We have sql 2005 standard installed as cluster on win2k3 cluster with 8
> GB memory on each node. We have one virtual sql server named instance on
> first node. Can we create one more instance on the same node and
> similarly 2 more instances on the second node. The cluster will be in
> active\active mode. Our reason for considering this is that we will be
> able to fail over one instance from the node 1 to node2 and node2 still
> have the 2nd instance running. Is this possible?
> Thanks
>
> --
> DeBe
> DeBe's Profile: http://www.dbtalk.net/m115
> View this thread: http://www.dbtalk.net/t297735
>
|||I will try to make my problem more clear.
I am talking about 4 different instances and they have 4 different
databases
connecting to 4 very different application.
so
Node1-->InsA-->DBa
Node1-->InsB-->DBb
Node2-->InsC-->DBc
Node2-->InsD-->DBd
So I guess my question is is it possible for just InsA to fail over to
Node 2. So that node 2 is now running with 3 instances and node 1 is
running with 1 instance?
My second question would be about the memory. If out of 8 GB Ram on
each node if 1 GB is assigned to os and 3 GB is assigned to each
instance (assuming 2 instances) and if the third instance from the node
2 fails over how will it get the memory for itself?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
|||You can have up to sixteen instances per cluster. Once the instances are
installed, you can set the preferred node order for each instance. That
determines which node is the normal host for this instance. Each instance
is completely independent from the other instances. You can move each
instance independently between nodes without affecting any other instance.
Just as on a stand-alone system with multiple instances, it is the DBA's
responsibility to make sure there are enough resources for all instances to
function properly. Note that SQL will not rebalance memory amongst multiple
instances. I.E. It will not make one instance give up memory so another one
can have it. You must plan ahead for the situation where one node hosts
more than its normal set of instances.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"DeBe" <DeBe.25uu4z@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25uu4z@.no-mx.forums.yourdomain.com.au...
> I will try to make my problem more clear.
> I am talking about 4 different instances and they have 4 different
> databases
> connecting to 4 very different application.
> so
> Node1-->InsA-->DBa
> Node1-->InsB-->DBb
> Node2-->InsC-->DBc
> Node2-->InsD-->DBd
> So I guess my question is is it possible for just InsA to fail over to
> Node 2. So that node 2 is now running with 3 instances and node 1 is
> running with 1 instance?
> My second question would be about the memory. If out of 8 GB Ram on
> each node if 1 GB is assigned to os and 3 GB is assigned to each
> instance (assuming 2 instances) and if the third instance from the node
> 2 fails over how will it get the memory for itself?
> Thanks
>
> --
> DeBe
> DeBe's Profile: http://www.dbtalk.net/m115
> View this thread: http://www.dbtalk.net/t297735
>
|||Just curious why you need so many instances. One instance can handle 4
DB's.
Tom
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Toronto, ON Canada
..
"DeBe" <DeBe.25uu4z@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25uu4z@.no-mx.forums.yourdomain.com.au...
I will try to make my problem more clear.
I am talking about 4 different instances and they have 4 different
databases
connecting to 4 very different application.
so
Node1-->InsA-->DBa
Node1-->InsB-->DBb
Node2-->InsC-->DBc
Node2-->InsD-->DBd
So I guess my question is is it possible for just InsA to fail over to
Node 2. So that node 2 is now running with 3 instances and node 1 is
running with 1 instance?
My second question would be about the memory. If out of 8 GB Ram on
each node if 1 GB is assigned to os and 3 GB is assigned to each
instance (assuming 2 instances) and if the third instance from the node
2 fails over how will it get the memory for itself?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
We have sql 2005 standard installed as cluster on win2k3 cluster with 8
GB memory on each node. We have one virtual sql server named instance on
first node. Can we create one more instance on the same node and
similarly 2 more instances on the second node. The cluster will be in
active\active mode. Our reason for considering this is that we will be
able to fail over one instance from the node 1 to node2 and node2 still
have the 2nd instance running. Is this possible?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
I think you have some misconceptions on how clustering works. The obsolete
naming of "Active/passive/Active..." is partially responsible for that.
Instances are installed to the cluster. You can have up to sixteen
instances on a cluster. SQL Server 2005 Standard Edition supports up to
two-node clusters. Nodes are the host computers for a cluster. An instance
can run on either node, but only on one node at a time. You cannot install
the same SQL instance to two nodes and have them both serviceing data
requests as if they were the same server. SQL Clustering is a failover
technology, not a scale-out technology. You can create two SQL instances on
a SQL 2005 two-node cluster and let one run normally on each node. During a
failure event one node can run both instances, with some performance
compromises of course.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"DeBe" <DeBe.25toh0@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25toh0@.no-mx.forums.yourdomain.com.au...
> Hi,
> We have sql 2005 standard installed as cluster on win2k3 cluster with 8
> GB memory on each node. We have one virtual sql server named instance on
> first node. Can we create one more instance on the same node and
> similarly 2 more instances on the second node. The cluster will be in
> active\active mode. Our reason for considering this is that we will be
> able to fail over one instance from the node 1 to node2 and node2 still
> have the 2nd instance running. Is this possible?
> Thanks
>
> --
> DeBe
> DeBe's Profile: http://www.dbtalk.net/m115
> View this thread: http://www.dbtalk.net/t297735
>
|||I will try to make my problem more clear.
I am talking about 4 different instances and they have 4 different
databases
connecting to 4 very different application.
so
Node1-->InsA-->DBa
Node1-->InsB-->DBb
Node2-->InsC-->DBc
Node2-->InsD-->DBd
So I guess my question is is it possible for just InsA to fail over to
Node 2. So that node 2 is now running with 3 instances and node 1 is
running with 1 instance?
My second question would be about the memory. If out of 8 GB Ram on
each node if 1 GB is assigned to os and 3 GB is assigned to each
instance (assuming 2 instances) and if the third instance from the node
2 fails over how will it get the memory for itself?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
|||You can have up to sixteen instances per cluster. Once the instances are
installed, you can set the preferred node order for each instance. That
determines which node is the normal host for this instance. Each instance
is completely independent from the other instances. You can move each
instance independently between nodes without affecting any other instance.
Just as on a stand-alone system with multiple instances, it is the DBA's
responsibility to make sure there are enough resources for all instances to
function properly. Note that SQL will not rebalance memory amongst multiple
instances. I.E. It will not make one instance give up memory so another one
can have it. You must plan ahead for the situation where one node hosts
more than its normal set of instances.
Geoff N. Hiten
Senior Database Administrator
Microsoft SQL Server MVP
"DeBe" <DeBe.25uu4z@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25uu4z@.no-mx.forums.yourdomain.com.au...
> I will try to make my problem more clear.
> I am talking about 4 different instances and they have 4 different
> databases
> connecting to 4 very different application.
> so
> Node1-->InsA-->DBa
> Node1-->InsB-->DBb
> Node2-->InsC-->DBc
> Node2-->InsD-->DBd
> So I guess my question is is it possible for just InsA to fail over to
> Node 2. So that node 2 is now running with 3 instances and node 1 is
> running with 1 instance?
> My second question would be about the memory. If out of 8 GB Ram on
> each node if 1 GB is assigned to os and 3 GB is assigned to each
> instance (assuming 2 instances) and if the third instance from the node
> 2 fails over how will it get the memory for itself?
> Thanks
>
> --
> DeBe
> DeBe's Profile: http://www.dbtalk.net/m115
> View this thread: http://www.dbtalk.net/t297735
>
|||Just curious why you need so many instances. One instance can handle 4
DB's.
Tom
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Toronto, ON Canada
..
"DeBe" <DeBe.25uu4z@.no-mx.forums.yourdomain.com.au> wrote in message
news:DeBe.25uu4z@.no-mx.forums.yourdomain.com.au...
I will try to make my problem more clear.
I am talking about 4 different instances and they have 4 different
databases
connecting to 4 very different application.
so
Node1-->InsA-->DBa
Node1-->InsB-->DBb
Node2-->InsC-->DBc
Node2-->InsD-->DBd
So I guess my question is is it possible for just InsA to fail over to
Node 2. So that node 2 is now running with 3 instances and node 1 is
running with 1 instance?
My second question would be about the memory. If out of 8 GB Ram on
each node if 1 GB is assigned to os and 3 GB is assigned to each
instance (assuming 2 instances) and if the third instance from the node
2 fails over how will it get the memory for itself?
Thanks
DeBe
DeBe's Profile: http://www.dbtalk.net/m115
View this thread: http://www.dbtalk.net/t297735
Subscribe to:
Posts (Atom)