This is part of a few memories from the founders of SQL Server Central, celebrating 25 years of operation this month.

"The site is down."

I got a phone call from one of the other founders around 9 am in Denver one day. At the time, I was working at a small startup, in a semi-private office with one other person. Most of the company knew I had a site on the side, and alternately cheered me on or celebrated hiccups, depending on the person.

I checked, and things were down. I had a side channel to telnet into the server, but couldn't access things. At this time, the site was hosted on a single server, running IIS and SQL Server, in a friend's basement. This was in 2002 or 2003, and broadband was a lot different then. No cable internet, and most other solutions were in the kbps range. I had ISDN at my house, but a friend had gotten into a trial from Sprint, giving him 3Mbps over a microwave link.

My office-mate was listening and watching. I said I was taking an early, and long, lunch. She chuckled and went back to her own tasks. I let my boss know and started driving, stressed out. After all, the site was growing, popular, and downtime could be a killer.

We've had a few outages over the years, but not many. This was one of the more stressful ones as I fought traffic on I-25, trying to get from the Denver Tech Center up to Westminster. I can't remember the weather, but I was sweating when I got to my friend's house. This was the before-times, when remote work was a rarity. Luckily, my friend had left me a key to get into his house, where I ran into the basement. Unable to get the server to respond, I rebooted it and crossed my fingers.

I'm sure a few of you have had the stressed-out feeling of waiting for a database server to restart, hoping that it comes back up cleanly. It did this time, and I don't remember what went wrong, but things were back up and running with no major issues. I was probably in the basement for less than 30 minutes, and I left, dreading the long drive back to the office. Almost two hours driving for a 30-minute fix.

I do remember thinking if this was going to be a regular occurrence, I might need a new job. Or a new place to host the server.

We weren't in the basement for too long. Revenues were increasing enough that I started to look at co-location facilities. My current employer had investigated quite a few when we set up our systems, and I reached out to a few contacts. I negotiated a half-rack at some point, moving our servers to a real facility. By that time, my employer had failed, and I got an F5 firewall as part of my severance, which fronted SQL Server Central for quite a few years.

Those were the days, with lots of CLI access to remote systems and a power strip where I could cycle them off and on. Those were skills I had to learn to avoid more drives at inconvenient times.