![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Hi all, I'm still reading around this and hoping to test in the lab, but I thought I'd get a quick pointer in the right direction first by someone who's done it before: I have a clustered SQL 2005. I want to break that and change to two standalone SQL to use setup mirroring instead, where then one node will be re-located to another building to meet some physical site DR objective. What's an overall process to do that conversion without reinstalling the primary node nor least interuption to apps that uses the clustered SQL's virtual name? In uninstalling the SQL an absolute must? Many thanks. |
#3
| |||
| |||
|
|
I would kick one node out, reinstall SQL on that node, move all the logins, settings, databases, etc to that server. Then I would rebuild the clustered node as standalone, install SQL on that machine, and move whatever databases over there that needed to go. There is no direct cluster to noncluster conversion. -- Geoff N. Hiten Principal SQL Infrastructure Consultant Microsoft SQL Server MVP "Elissa" <elissa.spam.lean (AT) gmail (DOT) com> wrote in message news:OYe9UGXdJHA.1132 (AT) TK2MSFTNGP05 (DOT) phx.gbl... Hi all, I'm still reading around this and hoping to test in the lab, but I thought I'd get a quick pointer in the right direction first by someone who's done it before: I have a clustered SQL 2005. I want to break that and change to two standalone SQL to use setup mirroring instead, where then one node will be re-located to another building to meet some physical site DR objective. What's an overall process to do that conversion without reinstalling the primary node nor least interuption to apps that uses the clustered SQL's virtual name? In uninstalling the SQL an absolute must? Many thanks. |
#4
| |||
| |||
|
|
Thank you Geoff for your pointers. One quickie, how do I retain the virtual SQL cluster name that has been configured into the applications after I break the cluster? Can I just use a CN record in the DNS? Cheers. ____ "Geoff N. Hiten" <SQLCraftsman (AT) gmail (DOT) com> wrote in message news:uhhq0nYdJHA.1704 (AT) TK2MSFTNGP05 (DOT) phx.gbl... I would kick one node out, reinstall SQL on that node, move all the logins, settings, databases, etc to that server. Then I would rebuild the clustered node as standalone, install SQL on that machine, and move whatever databases over there that needed to go. There is no direct cluster to noncluster conversion. -- Geoff N. Hiten Principal SQL Infrastructure Consultant Microsoft SQL Server MVP "Elissa" <elissa.spam.lean (AT) gmail (DOT) com> wrote in message news:OYe9UGXdJHA.1132 (AT) TK2MSFTNGP05 (DOT) phx.gbl... Hi all, I'm still reading around this and hoping to test in the lab, but I thought I'd get a quick pointer in the right direction first by someone who's done it before: I have a clustered SQL 2005. I want to break that and change to two standalone SQL to use setup mirroring instead, where then one node will be re-located to another building to meet some physical site DR objective. What's an overall process to do that conversion without reinstalling the primary node nor least interuption to apps that uses the clustered SQL's virtual name? In uninstalling the SQL an absolute must? Many thanks. |
#5
| |||
| |||
|
|
Yes, a CN record will work if youa re using the derault port (1433) otherwise you need a SRV record which allows you to specify a port number. -- Geoff N. Hiten Principal SQL Infrastructure Consultant Microsoft SQL Server MVP "Elissa" <elissa.spam.lean (AT) gmail (DOT) com> wrote in message news:OGobptddJHA.3520 (AT) TK2MSFTNGP05 (DOT) phx.gbl... Thank you Geoff for your pointers. One quickie, how do I retain the virtual SQL cluster name that has been configured into the applications after I break the cluster? Can I just use a CN record in the DNS? Cheers. ____ "Geoff N. Hiten" <SQLCraftsman (AT) gmail (DOT) com> wrote in message news:uhhq0nYdJHA.1704 (AT) TK2MSFTNGP05 (DOT) phx.gbl... I would kick one node out, reinstall SQL on that node, move all the logins, settings, databases, etc to that server. Then I would rebuild the clustered node as standalone, install SQL on that machine, and move whatever databases over there that needed to go. There is no direct cluster to noncluster conversion. -- Geoff N. Hiten Principal SQL Infrastructure Consultant Microsoft SQL Server MVP "Elissa" <elissa.spam.lean (AT) gmail (DOT) com> wrote in message news:OYe9UGXdJHA.1132 (AT) TK2MSFTNGP05 (DOT) phx.gbl... Hi all, I'm still reading around this and hoping to test in the lab, but I thought I'd get a quick pointer in the right direction first by someone who's done it before: I have a clustered SQL 2005. I want to break that and change to two standalone SQL to use setup mirroring instead, where then one node will be re-located to another building to meet some physical site DR objective. What's an overall process to do that conversion without reinstalling the primary node nor least interuption to apps that uses the clustered SQL's virtual name? In uninstalling the SQL an absolute must? Many thanks. |
![]() |
| Thread Tools | |
| Display Modes | |
| |