Ah. I had a little downtime recently … or … three days of downtime, actually. It seems as though I did ‘something to something’ a long time ago that I didn’t test, so when I rebooted the server it decided to curl into a ball and rock back and forth.
Oops!
Worse, the last (and only) daily and weekly backups did nothing. I was screwed.
I thought it would be straightforward to get nginx, php, mysql running again, considering this is the third time I’ve set it up. How wrong I was. My first attempt was met with bad gateway errors, it didn’t seem like fcgi processes were spawning. Even the generally quiet mysql was complaining about about sockets.
My next rebuild I tried to get PHP-FPM working, but I quickly realized that I had no idea what I was doing and that patching and building php would probably be the death of me later on.
Three tries later and I was back up and running:
- Initial Setup
Slicehost articles: Intrepid setup articles (one, two)
I’m running 9.04 (Jaunty Jackalope) but it all worked splendidly. - Installing nginx
Slicehost articles: nginx setup from source (one, two, three)
I grabbed the latest stable release of nginx, not the one used in the article. - PHP + nginx
Then came the dilemma of how to spawn FCGI PHP processes. Having completely forgot how I did it previously, I decided to go a route I know I didn’t take — using the lighttp fcgi launch script thing. (this article helped immensely with that).
Overall, things appear to be running smoothly now. Fantastic.
I’ve had it with server admin for the time being.