Lighttpd server is stopped

Question

I have a Lighttpd server plus mod_fastcgi. And today I had Internal Server Error 500. I've checked my error log and it goes like this:

2010-04-22 22:59:14: (server.c.1464) server stopped by UID = 0 PID = 3332
2010-04-22 22:59:15: (mod_fastcgi.c.1768) connect failed: No such file or directory on unix:/tmp/php.socket-5
2010-04-22 22:59:15: (mod_fastcgi.c.2956) backend died; we'll disable it for 5 seconds and send the request to another backend instead: reconnects: 0 load: 1
2010-04-22 22:59:15: (mod_fastcgi.c.2709) child died somehow, waitpid failed: 10
2010-04-22 22:59:15: (server.c.1464) server stopped by UID = 0 PID = 3332
2010-04-22 22:59:15: (server.c.1464) server stopped by UID = 48 PID = 1385
2010-04-22 22:59:15: (server.c.1464) server stopped by UID = 48 PID = 1385
2010-04-22 22:59:15: (server.c.1464) server stopped by UID = 48 PID = 1385
2010-04-22 22:59:15: (server.c.1464) server stopped by UID = 48 PID = 1385

What to do to find out what can be cause of this?

Answer 1

Some possible reason:

1. /tmp partition is full
2. /tmp has wrong permission code, which forbidden the /tmp/php.socket-5 to be created

Answer 2

You can try to run the server with strace and look which syscall failed. Make sure that the server does not detach from the terminal (=does not go in daemon mode, I didn't find if lighthhtpd does have a command line switch for that, it has a -D switch but I can't see if it has the desired effect), else you won't see the output for the failing process.