Because Microsoft doesn’t know how you are going to shake this stick.
Each SCR source can have multiple targets. A SCR source can be a SCC, a CCR, or a standalone server. A SCR target can be a SCC, CCR, or standalone server. A given AD account can only be tied to one of them at a time, even if there are multiple databases containing multiple copies of the same mailbox! So…which source? Which target? Which lag set? How quickly to fail over? How many logs do I apply? Do I update DNS? With choice comes complexity. (Believe it or not, I was just involved in a discussion with a set of MSFT engineers this morning taking your side of this argument – that it’s too complicated. So….I agree with you. And I believe that it will get better. You have many advocates such as me, Neil, and William all trying to keep things workable for the SMORG.) From: John Bowles [mailto:john_bow...@yahoo.com] Sent: Wednesday, March 11, 2009 2:06 PM To: MS-Exchange Admin Issues Subject: Re: Exchange 2007 SCR -- Hardware? Is there any reason that MS hasn't made this more of an automated process without going into PS and ripping through the command line while you're Exchange boxes are down? I don't see the benefit in SCR when it comes up time unless you have some guy on staff making 6 digits that knows a great deal about PS. If you take an average ExAdmin and throw E2K7 and they require site resiliency.. I can see an admin drowning with all the command line info you need to remember to move everything over. The last time I checked the whole idea around "Windows" was to make everything easier for people.. it seems they're going in the opposite direction with their new puppy. Just my opinion. _____________ John Bowles _____ From: Michael B. Smith <mich...@theessentialexchange.com> To: MS-Exchange Admin Issues <exchangelist@lyris.sunbelt-software.com> Sent: Wednesday, March 11, 2009 11:30:32 AM Subject: RE: Exchange 2007 SCR -- Hardware? See, I fall directly into the “database portability” camp. Who wants to do a /RecoverCMS when you can just do a few “set-storagegroups”, “set-mailboxdatabases”, ”mount-database”, and “move-mailbox –configurationonly” --- and I can script the entire thing ahead of time! The only downtime is DNS TTL across sites and you have the same issue with single-node clusters. (Granted, this presumes Outlook 2007 or higher in the environment.) From: Neil Hobson [mailto:nhob...@gmail.com] Sent: Wednesday, March 11, 2009 11:20 AM To: MS-Exchange Admin Issues Subject: RE: Exchange 2007 SCR -- Hardware? Yep, that’s been the case with all of my SCR deployments to date. The only slight difference with some designs is that the standby cluster is sometimes just a single node cluster (initially, anyway) From: Michael B. Smith [mailto:mich...@theessentialexchange.com] Sent: 11 March 2009 14:54 To: MS-Exchange Admin Issues Subject: RE: Exchange 2007 SCR -- Hardware? When I recommend someone to deploy SCR, I recommend identical hardware. The general goal of SCR is fault tolerance and site resilience. The SCR hardware can’t take over if it can’t handle the load… IMHO. YMMV. From: 8400...@gmail.com [mailto:8400...@gmail.com] On Behalf Of jond Sent: Wednesday, March 11, 2009 10:43 AM To: MS-Exchange Admin Issues Subject: Exchange 2007 SCR -- Hardware? For the people here using SCR in Exchange 2007 SP1, what kind of hardware are you using relative to your production exchange boxes? Are you finding that as long as you have enough hard drive space, that processor and ram don't really matter or did you simply decide to run identical hardware on both production and on the SCR server? Thanks in advance, Jon ~ Ninja Email Security with Cloudmark Spam Engine Gets Image Spam ~ ~ http://www.sunbeltsoftware.com/Ninja ~
