I have read a thread here on how to properly shutdown a SQL 2000 and 2005
cluster and wondering if there is any difference with SQL 2008 cluster. We
have a two node active passive cluster with a SAN. I will list my steps here
and please advise if it is not right. We need to shut down both nodes
because of network reconfiguration.

1. Take the SQL resource offline and set the start up type from Manual to
Disabled on all nodes
2. Stop the windows cluster service on all nodes and set the startup type
from automatic to disabled.
3. shut down Primary node (active)
4. shut down Passive node
5. Shut down the SAN
6. Do the network maintainance
7. turn on SAN
8. Boot primary node server
9. Boot passive node server
10. Start the windows cluster service on the primary node and set the startup
type from disabled to automatic
11. Start the windows cluster service on the passive node and set the startup
type from disabled to automatic
12. Test the failover
13. Start the SQL resource on the primary server and set the start up type
from Disabled to Manual.
14. Set the startup type on the passive server from Disabled to Manual.
15. Test the SQL failover

OD

--
Message posted via SQLMonster.com
http://www.sqlmonster.com/Uwe/Forums...ering/200910/1

The steps for controlled shutdown and startup for a SQL Cluster do not
change for SQL /Windoes2008 with one exception.

For Windows 2008 clusters with three or more , you need a majority of nodes
+ quorum. The Quorum disk is no longer the absolute ruler, it is just a
single vote.

--
Geoff N. Hiten
Principal SQL Infrastructure Consultant
Microsoft SQL Server MVP



"OceanDeep via SQLMonster.com" <u46587@uwe> wrote