How does this command with which to "refresh the syslog process" work?

Question

On http://www.cs.newpaltz.edu/~easwaran/Resources/commands.html the following command is listed as being meant to "Refresh the syslog process":

kill -1 `cat /var/run/syslogd.pid`

I have seen this command in real-life use on an embedded device but I cannot wrap my mind around why someone would use it and how it works.

I understand the single parts this is made up off, but e.g. why is the cat subshell killed using a HUP signal is not something which makes sense to me.

Answer 1

syslogd would appear to do two things for the convenience of system administrators:

(a) If it is running, it posts its own pid (process id) in a well-known file. When it exits, it will attempt to remove this file.

(b) When it receives SIGHUP, it takes that as a hint that it should re-examine its configuration, because something might have changed recently. This is something of a convention: e.g. Unix cron daemons needed to be sent SIGHUP after a contab file was edited, to ensure any changes became effective.