October 20, 2010 at 1:54 am
Hi all,
I'm pretty green when it comes to sql server clustering so I wondered about this scenario (which we actually have run into at work):
Let's say you have an alias, Alias, for sqlserver cluster comprising of 2 servers, Server1 and Server2, like this:
ALIAS ----+-----> Server1
|
|
+----> Server2
If a programmer where to connect directly to Server1 without going via the Alias and change data, how does Server2 stay in sync in case of a fail-over? Does the cluster software read the logfiles on the "live" server or something like that?
Thanks!
Dennis
October 20, 2010 at 2:17 am
A cluster uses shared disks. There's only one copy of the database files on the shared storage and only one active instance of SQL (unless we're talking about multi-instance clusters)
Gail Shaw
Microsoft Certified Master: SQL Server, MVP, M.Sc (Comp Sci)
SQL In The Wild: Discussions on DB performance with occasional diversions into recoverability
October 20, 2010 at 2:38 am
OK, yes, that makes sense. For some reason I thought that each instance had its own copy of the data on its "own" disks...thanks!
October 20, 2010 at 5:27 am
when connecting to a cluster you connect using the virtual sql instance name, not the name of the physical cluster node. The virtual name connects you to whichever node is actually supporting the cluster. An alias should not be necessary.
---------------------------------------------------------------------
Viewing 4 posts - 1 through 3 (of 3 total)
You must be logged in to reply to this topic. Login to reply