It seems that all I write, when I write, is about bad customer experiences. I’ve got another one to detail today. But before I do so, I’d like to make the point about why I write these reviews. First, I can get this aggravation off my chest. I’ve had to take action that I didn’t really want to. The reason why was entirely preventable. And that’s my second point. This is a teaching moment. When something annoying happens and I feel compelled to write about it, it’s a reminder to both myself and the world that it could be done better in the future. Hopefully we all learn from that.
I purchased a dedicated server at the end of July from the very large hosting provider 1&1. The server was for one of my best clients so it had to be right. Without boring all the details, let’s just say that it’s purpose was to host 1K+ email accounts and be a hot-backup and fail-over machine if the primary server goes offline.
After bringing it online and loading it up, it ran well for a while. However, by mid-August the server began freezing and/or crashing. Their technical support department was remarkable little to no help. Their attitude was “it’s your server, it’s your problem.” I could accept that judgement if I was running some funky custom set-up. However, it was off the shelf CentOS 6 with all the patches and Plesk 12. Nothing strange here.
Eventually I gathered enough evidence to show that there was a hardware problem. I reported this to a sympathetic tech support rep. The next day, without prior notice, engineering took my server offline and replaced the hardware. Then the server wouldn’t work because the MAC address had changed. Once again technical support offered little help. Eventually I was able to get the server repaired by the grace of a friend of mine (and sysadmin of Khoza Tech) and carry on. However, the crashes didn’t end there.
Once, after logging off from using the 1&1 web admin panel to restart the server, I was presented with a feedback form. I leapt at the chance to write a detailed note explaining my frustration with the server. The result of that feedback? Nothing. No contact from 1&1. No acknowledgement whatsoever.
Ultimately I had to move on. By early September I had documented ten separate occurrences of mysterious crashes. I’m a patient guy but enough’s enough. My dream of geographically diverse servers was not to be. Instead I had to settle for a slightly more expensive solution in Fremont. While that’s still probably far enough away to survive just about any disaster short of an asteroid strike, I’m still disappointed. I’ve been doing business with 1&1 for years. I recommended a dedicated server from them to a good client. That server has been rock solid for years – never a crash or failure. Perhaps the hardware they were using for the server I picked is just not reliable. After all, I’d never heard of using multi-core Atom processors for generic hosting before. Either way, the real issue is a failure by 1&1 to take care of me as a customer. I would have gladly locked in with them for years but they didn’t even acknowledge me after I had gone out of my way to get their attention and tried to get things right. I hope that they’ve got other satisfied customers because they lost this one.