Wow The only time I had this happen to me.... I lost a drive and replaced it. As the array was rebuilding, someone kicked the plugs out of the server. I booted it back up and lost the stripe set. I had to restore 200 gig worth of data (thank goodness it wasn't Exchange). Moral of the story - don't let dingbats in your server room - no matter what the boss says......
Thanks Russ -----Original Message----- From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] On Behalf Of Matthew Joyce Sent: Tuesday, January 18, 2005 10:38 PM To: Exchange Discussions Subject: DR Plan Hi, Firstly, could someone direct me to the archive of this list please, I know the subject of DR is a recurrent one. Last week was truly awful for me; we lost email for a week, something I would never have thought would happen. Basically a drive failed and corrupted a RAID5 volume; I had heard of it happening, but never experienced it before. It's very hard to explain to bosses that we have RAID to provide resilience, but sometimes it doesn't work. It took a long time to declare the volume dead (waiting for new drives, waiting for engineers). Building the new server was problematic, restore system state just didn't work. It had been the only DC on site which complicated things. Restoring the stores was also not smooth and took along time, then the stores would not mount and all that took ages. After logging a call with Microsoft, who insisted I do everything I had already did again and told me 'that should have worked'. (ring them at 0800, they will call back in 4 hours, ring them at 2000, they have gone home) Our upstream mail handler had their servers looping our incoming mail and causing it to bounce with 'too many hops'. When our server came back up the mail which was being store by other upstream handlers, rushed in and went nowhere. /me shudders Anyway, we're back now, and I want to write the mother of all DR plans while the incident is fresh. Essentially there was far too much learning going on, and not enough knowing. The awful truth is we had planned to start DR testing this month. I'm sitting here reading articles and making note about how our system is setup, I just don't know what should be included. Can anyone give me some pointers ? Are there any template I can use to create useful documents. Our setup now is a single Exchange 2003 Enterprise server, 2 sites, 2 DCs (1 on each site) The server now has two mirrored RAID arrays, logs and system on one and stores on the other. I backup the Stores+flush on Monday and the logs the rest of the week. So if this server goes 'pop' is my procedure something like : don't have contact details only in outlook know where software and licenses are know where ssl certs are make sure isps are holding mail and for how long organise suitable hardware setup arrays reset computer account install os same name, different IP (to stop mail coming in) join domain install exchange /disasterrecovery restore stores and logs connect account to mailboxes check internal mail install av and as reset ip check mail flow make backup Clearly this need fleshing out, but is is essentially correct ? I know things go wrong and being able to cope with those moments is what is going to make a DR plan better, but is this the basic best case scenario ? Is it worth perhaps having a linux box (debian) as a smarthost (configured not to deliver by default) to catch incoming mail while issues are resolved ? what's the score with changing the IP address to control mail flow is this sane ? We have about 140 mailboxes, which may not seem like many but everyone is important. I consider myself lucky, our staff were all very patient, tolerant and surprisingly understanding, I've been in IT for 16 years, and to lose incoming mail the way we did, utterly mortifies me. Thanks Matt Joyce Children's Cancer Institute Australia http://www.ccia.org.au _________________________________________________________________ List posting FAQ: http://www.swinc.com/resource/exch_faq.htm Web Interface: http://intm-dl.sparklist.com/read/?forum=exchange To subscribe: http://e-newsletters.internet.com/discussionlists.html/ To unsubscribe send a blank email to [EMAIL PROTECTED] Exchange List admin: [EMAIL PROTECTED] To unsubscribe via postal mail, please contact us at: Jupitermedia Corp. Attn: Discussion List Management 475 Park Avenue South New York, NY 10016 Please include the email address which you have been contacted with. _________________________________________________________________ List posting FAQ: http://www.swinc.com/resource/exch_faq.htm Web Interface: http://intm-dl.sparklist.com/read/?forum=exchange To subscribe: http://e-newsletters.internet.com/discussionlists.html/ To unsubscribe send a blank email to [EMAIL PROTECTED] Exchange List admin: [EMAIL PROTECTED] To unsubscribe via postal mail, please contact us at: Jupitermedia Corp. Attn: Discussion List Management 475 Park Avenue South New York, NY 10016 Please include the email address which you have been contacted with.
