In hindsight, perhaps this wasn’t the best time for me to be launching a blog. Then again, in this line of work there’s never a good time for anything. My company acquired a new office about a month ago, and we just spent the past weekend bringing them onto our network. As these maneuvers go, this one seemed to give us more trouble than usual.
Our first order of business was connecting the new office to our WAN. We knew we were in trouble fairly early Friday evening when we ran into trouble bringing a new router online. It turned out that our telecomm vendor provided our router supplier with an incorrect config. If that wasn’t bad enough, when we attempted logging in ourselves to fix the config, the password we were given didn’t work. At that point of the evening, all of our usual contacts at both the provider and our hardware vendor had disappeared. The folks we were able to reach were all telling us that we’d have to wait until Monday.
Thanks to the persistence of my staff, a little help from Cisco tech support, and an assist from a sympathetic tech on our provider’s help desk – we managed to bring the connection to life a little after midnight. We were now 6 hours behind schedule, but at least we were connected. We quit for the evening figuring that the next step, creating a new Windows child domain was going to be easy.
Wrong again. The next day found us once again struggling as we attempted to run DCPromo on the local file server. After 5 or 6 more hours of struggle and a few hours on the phone to Microsoft, we finally found the culprit. Our new server was running Windows Server 2003 R2, and we had never run ADPrep for R2 back at the root of our forest. Ultimately it was a relatively simple fix, but now we were well over 10 hours behind schedule, and all our dreams of taking a Sunday off in sunny Florida had evaporated. We still had plenty left to do, but we worked through the day on Sunday and managed to finish up sometime after 1:00 am on Monday morning.
When Monday morning rolled around, people were connected and things were working fairly smoothly. At some point, somebody asked my why we didn’t document these things. Actually, we do, but the problem with documenting issues on something like this is that no two of these moves are the same. It’s still worth doing, but rarely do the same issues arise twice. We do occasionally add new offices or move existing offices, but it’s just not something we do with any great frequency. One issue which has bitten us more than once is file moves, especially where the Macs use strange characters in file and folder names. That one got us again this time, but I’ll leave the details of that, along with our other lessons learned for my next post.