0

We are wanting to phase out our reliance on the server internal drives and start running things from an iSCSI LUN (for various uninteresting reasons). The first few weeks of this plan mainly involved running up Exchange LCR to an iSCSI drive and watching it run to test the stability of the connection. This has been happily replicating for a few weeks now without a single error, so we are ready to swap over, that is, we want to start running the active database from the iSCSI drive and replicate it to a passive database on the internal drives.

Long story short, I've been reading up - or rather, trying to read up - on the procedure behind this. Unfortunately all the available documentation seems to come with the assumption that you are recovering from a failure. That's understandable, I get how it's the main point of the feature, but it leaves me a little paranoid about the situation. I want to make sure I don't create a disaster by using disaster response steps in a disaster free situation essentially.

The most concise steps I have found are at http://www.exchangeinbox.com/article.aspx?i=138 under the section "Restoring a local Continuous Replication Database", and gosh don't they make it look simple!? That "Elapsed Time: 0:03" in the screenshot has my hopes right up that this is something we can knock out right on 5PM and still get out the door by 5:15. Am I being naive?

So essentially, what I want to confirm is that by following this process:

  • We are just changing the configuration of exchange, such that it looks for the active database in the new location (this as opposed to copying the whole database from place to place or something similarly time consuming)
  • We should know within a fairly short time whether we have been successful or not (that is to say, I don't have to leave some long running command overnight and cross my fingers that mail will be up in the morning)

1 Answer 1

0

To answer this myself - we did this last night and indeed it took 2 minutes - breezy. The search indexer had to run a full crawl - this was all I did not anticipate - which caused the disk reads/sec to max out for a while, but while this slowed local delivery, it was not too extreme (messages I sent to myself from my exchange account arrived within 4 minutes, messages from my gmail to exchange still came in <10 seconds). All in all, a painless process.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .