Not long ago, ChristopherCantwell.com had gotten too popular for our shared hosting provider to handle. There were frequent outages during traffic spikes, and our provider, HostGator, was less than helpful. The day before Thanksgiving, they locked out the account and stopped responding to our support requests. There were several days of continuous downtime on all my sites due to this, and it was loud and clear that I could no longer trust my sites to HostGator.
I acquired a dedicated server from Ezzi.net, who is now also a sponsor of this website. I know these guys, I used to work for them as the Operations Manager. One of their techs is an original developer of the cPanel Web Host Manager, and they are very helpful. Things are much improved, and we won’t be having any more outages due to traffic spikes. I also signed us up for CloudFlare, which should further reduce server load and page load times.
ChristopherCantwell.com is not the only site that I run though. I have several, some more popular than others, some that you would recognize, some that you would not, and some which I hope become very recognizable in the future. During the transition between providers, something on one of those sites became corrupted, and is causing a process to spiral out of control and consume all server resources. Worse than that, it seems to have this habit of doing so while I’m asleep or on the radio, so I can’t troubleshoot it while it is ongoing. That was the cause of the outage last night, which was resolved by Ezzi before I got out of Free Talk Live. I troubleshot and monitored things until around 6:30am when I finally went to bed, and everything was fine. By 7:30am, the site was down again and I didn’t realize this until I got up around 11:30am.
The solution to this is fairly simple, but time consuming and tedious. I won’t bore you with the details except to say I have to further isolate each of the websites on the server so they don’t effect each other as much. When we’re done, one site screwing up won’t bring down the entire server. I’ll be working on this today and tonight, and hopefully we’ll have this all sorted out by tomorrow morning.
In the meanwhile, if you try to visit this site and encounter an error, please just wait a little while and then refresh the page. I have monitoring software running, and I’ve recruited some assistance in the event of an outage while I’m not available. Hopefully by tomorrow we can forget this ever happened.