Interesting thing happend today. At about 5 PM I was posting the last article on the site when I noticed that the server started becoming unresponsive. I was at work so there wasn’t much I could do in terms of investigation. See, our network nazis started blocking port 22 a couple years ago when using it for a tunnel. They told me it was a “security risk” but I find that hard to believe. My theory; they couldn’t “see” what was going in and out of the tunnel or what it was being used for (Jabber mostly… they block port 5222 & 5223 as well) so they shut it down. Yet they still allow FTP. I don’t under stand it.
Anyway, back to the issue at hand. So it appeard to be that the server was hung up on something. I checked my other blog and it was working perfectly. It’s not using ExecWrap + PHP like this blog. It’s using Typo, a Rails application. Weird.
When I got home, I logged in and check my memory usage. Seemed about average.
Total: 239 MB Used: 127 MB Free: 112 MBSo, I restarted Lighttpd and opened up a browser. Nothing. Server seems unresponsive. I restart it a couple more times. Still nothing. Hmm. I’m stumped at this point. So, I stop lighty, manually kill all the php-cgi processes and ruby processes and start lighttpd back up. Everything is normal now. Good. But, what caused the issue? I looked at all the log files I could think of. Couldn’t find anything anywhere.
I assume it was PHP that was the culprit since once I killed that, the server came back to life. I guess it could have been ExecWrap, but that stuff doesn’t really get logged anywhere (that I’m aware of).
I’ll have to keep my eye on it and see if this starts happening more frequently. So far, everything else has been running smoothly.