I'm trying to rescue our last minute DR from turning into a total cluster****
Here's the scenario. We have our DR server site in Cincinnati. Our DR client location is in Columbus. Our management will not allow us to move the DS server to the Columbus site, nor our client location to Cincinnati. The latency between the sites means that building an actual image, or 50 in our case, over the connection between the two is impossible in the time frame we have.
To add to this, we dont know what the client hardware will be. Theyre going to be provided to us from our parent company who does not share our exact environment. It may be machines that we already have images for, but its just as likely that they are machines that we do not have in our environment. All we know is that they will be Dells.
I wont even go into what the current plan is, its so backwards and nonsensible. I've had a couple of ideas but as of this moment they are untested. It's not actually my responsibility this year for our DR planning, but I feel I have to step in and try to help. If anyone has any suggestions on an approach, I would be so grateful... I'll mail you a cookie.
My current idea is to take on of our images via portable HDD or DVD etc, share it out on one of the client machines in Columbus. We could run the share off of a prebuilt laptop, or build the first machine manually. If the hardware is not already accounted for, we'll need to drop appropriate drivers and HAL files into the image using image editor.
Unless I'm mistaken, we should be able to have the machines boot to PXE, pull down the linux automation client, modify the mapping of the image location to point to the local share point with the modified images and bypass the slow pipe between our actual DS server and the clients.
Are there any major snags I havent seen here? I'll be taking it to the lab as soon as possible, but input is greatly appreciated.
Thanks