itsonlyme4
asked on
recovering a sql server
Running multiple sql server 2005 instances on windows 2008.
Our Server guys changed the way that they back up the server (at the server level) and excluded all .ldf and .mdf files.
Needless to say when we recently went through a disaster recovery drill I had to uninstall and reinstall sql server (all intstances) before I could begin restoring databases.
I am wondering if I had made copies of the master, resource, msdb and model .mdf and .ldf files for each instance and then copied them into their correct directories for the DR drill if i could have avoided having to reinstall SQL Server? I don't have a server to test this on which is why I am asking....
How do others backup their servers hosting sql server?
Our Server guys changed the way that they back up the server (at the server level) and excluded all .ldf and .mdf files.
Needless to say when we recently went through a disaster recovery drill I had to uninstall and reinstall sql server (all intstances) before I could begin restoring databases.
I am wondering if I had made copies of the master, resource, msdb and model .mdf and .ldf files for each instance and then copied them into their correct directories for the DR drill if i could have avoided having to reinstall SQL Server? I don't have a server to test this on which is why I am asking....
How do others backup their servers hosting sql server?
So in the event of disaster, i can restore my backups from Full + Diff + My log backups.
<<I am wondering if I had made copies of the master, resource, msdb and model .mdf and .ldf files for each instance and then copied them into their correct directories for the DR drill if i could have avoided having to reinstall SQL Server?>>
Including system databases's restoration as a part of your system recovery strategy may reduce significantly your system restore time since you don't have to restore the logins, job information and additional system information that can not be restored through user database restoration.
Your system recovery process must also take into account the fact that the environment is clustered or not and network connectivity issues. Connectivity wise, I suggest you use an alias layer to be able to redirect all of your applications quickly to any instance no matter how it could be renamed. In a clustered environment make sure to document well all system cluster considerations: DTC creation, ressource creation, account service authentication issues.
Hope this helps...
Including system databases's restoration as a part of your system recovery strategy may reduce significantly your system restore time since you don't have to restore the logins, job information and additional system information that can not be restored through user database restoration.
Your system recovery process must also take into account the fact that the environment is clustered or not and network connectivity issues. Connectivity wise, I suggest you use an alias layer to be able to redirect all of your applications quickly to any instance no matter how it could be renamed. In a clustered environment make sure to document well all system cluster considerations: DTC creation, ressource creation, account service authentication issues.
Hope this helps...
ASKER CERTIFIED SOLUTION
membership
This solution is only available to members.
To access this solution, you must be a member of Experts Exchange.
Daily full backups of all System Database (Exclud tempdb).
Full backup of User database on Week end(After index Defragment).
Daily Diff backups of user database
Every 1 hr log backup(depends on the transactions).