What are the chances !!! exchange down !

[bitslice]

plus with 300+ users wouldn't like to be the one to manually export the user's PST files !!

restore the data to another server (anything will do)
use EXMERGE to export the mailbox data, on masse

drop the PST's into their home drives and frig the profile to add a personal folder.

(not ideal, just a quick way of getting them running on slow hardware if you have NO alternative)

having the mailboxes archived at the point of failure helps seperate the old data from the new mail.
ie while you are recovering their old data, a new server needs to be up and running to host the new mail arriving. (I presume the ISP is queuing it for you ?)



what is this chaps DR plan for this situation ?

.

 

[stoofa]

someone correct me here,
...swapping in a new Dell raid card may not fix the array

(hope I'm wrong, swappable RAID hardware is one of the plus sides of having a HP I've been told)

.

It will be one of the onboard Dell PERC RAID controllers which really aren't that bad.
Most Dell servers have the controller built onto the motherboard and then you can "activate it" so to speak by buying the add-in card.
However the add-in card is probably not going to be at fault, if anything it's going to be the onboard RAID controller itself and on a Dell server that is out of warranty.
As with all things RAID - so long as you use the same controller 9/10 your data will be intact if you swap the controller, it just needs to be the same make and model.
With Dell controllers you might get away with a newer controller being able to read the array created from an older controller.

It'll be times like this when you'll be happy the buck doesn't stop with you and that you've got a boss who will come out far worse than you here.
It's nobody's fault the hardware has failed - it happens and happens more than some people realise.

What seperates the good IT people from the bad ones is the DR procedure at this point and just how long the company remains without e-mail.
Last serious Exchange server issue we had (2x HD failing in a RAID 5 array) we were down for 8hrs.

 

[bitslice]

thanks for the correction Stoofa.
The add in card is just a chip key I think ?

Then there is always the problem of matching the firmware on replacement components, even Dell backplanes have bloomin firmware...
The last (Hardware) Exchange problem I had was solved by running over to the Computer company next door to buy another SCSI drive. Phew.

I'd prefer to fix the hardware, but if it's likely to be down for more than a few hours I'd have to reroute the incoming mail to the other server and sort the rest out as I go. (Having a gateway email archive and a mailbox backup to disk helps a lot.)

I'd reckon in this case a hardware fix (new motherboard ?) is well over a day away, and they might need some kind of email access in the meantime ?
If they have no restore server, then they will need to get a bit creative with whatever is lying around. That's why I suggested reducing the store size before bringing a temp replacement box online - but yeh, it's not ideal!

OP: After it's fixed would be a good time to review what else could go wrong.
Management are more likely to cough up for better hardware when it's still fresh in their minds.


.

 

[Ev0]

and the server is out of warrenty.

oh dear, if I've learnt anything from our server chaps is make sure your hardware is supported in case of part death. Was a godsend when one of the disk controllers went on one of our mail servers, and when something on our eva packed in.

 

[Slinwagh]

.........what is this chaps DR plan for this situation ?

.

Was that a tongue in cheak remark, your so cruel!

Whats the latest then on this?

An old IT guy said to me once that "You are only as good as your last backup" it might sound daft but it has helped me in a couple of situations.

 

[bitslice]

nah, I'm just curious.
he might actually have one, it's the kind of thing insurance companies want to know these days.
anyway I like hearing about other ways of doing stuff, half of my plans are cribbed off other IT people

OP: any update yet ?
.