![]() | |
![]() |
| | Thread Tools | Display Modes |
#31
| |||
| |||
|
|
"John Toner [MVP]" <jtoner (AT) DIE (DOT) SPAM.DIE.mvps.org> wrote in message news:OqoEHJfPJHA.3884 (AT) TK2MSFTNGP02 (DOT) phx.gbl... Option 1: Move one node to your Primary site, so you have 3 at Primary, 2 at secondary. The benefit of this is that your primary site will remain operational during a communications issue between the sites. The downside is that in the event of a site failure of the Primary site, you will need to manually bring the applications online at the remote site using the /forcequorum option. See MSKB 258078 for more details. Option 2: Use a disk based quorum. In a normal disk quorum cluster, the site where the quorum resides will typically survive the site failure. Benefits are that either site would survive this failure. Downside is that there are few geo-cluster solutions that support a shared disk quorum model. I know that EMC SRDF/CE is one that does support a quorum disk model in a geo-cluster. Option 3: Extend the MNS to include one instance in a third location and then configure the resources so they will not run on the node in the third site. Better yet, make the third site a File Share Witness. Of course, in 2003, you need to configure a VLAN to stretch over all three locations. -- Russ Kaufmann, MVP, MCSE: Messaging and Security, MCT, MCITP, MCTS and other stuff ClusterHelp.com, a Microsoft Certified Gold Partner Web http://www.clusterhelp.com Blog http://msmvps.com/clusterhelp |
![]() |
| Thread Tools | |
| Display Modes | |
| |