![]() | |
![]() |
| | Thread Tools | Display Modes |
#1
| |||
| |||
|
#2
| |||
| |||
|
|
Gurus, Running SQL Server 2005 SP2 on a cluster (two Windows Server 2003 SP2 nodes, SQL instance running on each, active/active design). A hardware refresh cycle has arrived and two new sparkling servers are online in our data center to replace the old ones. I didn't request this, they just showed up!!! Nice, huh? Anyway, what would be the best plan of attack in migrating the SQL instances off of the old cluster and onto the new? The new OS will be Windows 2008. -- Spin |
#3
| |||
| |||
|
|
Interesting timing - I will be leaving for work in about an hour to begin our migration to a brand new cluster. For me, the simplest option is a straight backup/restore, since our new servers are also on a new SAN. We are retiring the older EVA 5000 and replacing with Netapp SAN. Here is the process: - Install SQL Server 2005 SP3 on new servers as new cluster instance - Run the application install for the current version of the application (builds default databases and CLR components) - Restore backups of current system over new databases - Test the new system (I/O tests, connectivity, application, etc...) The above has already been done - now, tonight we are going to do the following: - Disconnect application servers - Put databases on current system in restricted mode - Backup current databases - Restore databases on new system/instance - Modify application server connections - Note: this could also be handled by updating DNS if needed - Release system to I/S team to test and validate Back-out plan is real simple, if everything goes bad - bring old system back online, modify connections and release to users. Identify the problems and rescheduled. Jeff |
#4
| |||
| |||
|
|
Gurus, Running SQL Server 2005 SP2 on a cluster (two Windows Server 2003 SP2 nodes, SQL instance running on each, active/active design). A hardware refresh cycle has arrived and two new sparkling servers are online in our data center to replace the old ones. I didn't request this, they just showed up!!! Nice, huh? Anyway, what would be the best plan of attack in migrating the SQL instances off of the old cluster and onto the new? The new OS will be Windows 2008. -- Spin |
#5
| |||
| |||
|
|
So how did everything go? Any lessons learned? "Jeffrey Williams" <jeff.williams3188 (AT) verizon (DOT) net> wrote in message news B6ACDC0-B14E-494E-95C8-ECF9F2354AC4 (AT) microsoft (DOT) com...Interesting timing - I will be leaving for work in about an hour to begin our migration to a brand new cluster. For me, the simplest option is a straight backup/restore, since our new servers are also on a new SAN. We are retiring the older EVA 5000 and replacing with Netapp SAN. Here is the process: - Install SQL Server 2005 SP3 on new servers as new cluster instance - Run the application install for the current version of the application (builds default databases and CLR components) - Restore backups of current system over new databases - Test the new system (I/O tests, connectivity, application, etc...) The above has already been done - now, tonight we are going to do the following: - Disconnect application servers - Put databases on current system in restricted mode - Backup current databases - Restore databases on new system/instance - Modify application server connections - Note: this could also be handled by updating DNS if needed - Release system to I/S team to test and validate Back-out plan is real simple, if everything goes bad - bring old system back online, modify connections and release to users. Identify the problems and rescheduled. Jeff |
![]() |
| Thread Tools | |
| Display Modes | |
| |