A couple of things to note
The script still makes the assumption of having a config in /etc/osquery/osquery.conf however it now checks to see if there is the default example config in /usr/share/osquery/osquery.example.conf and alerts the user that it's using the default.
To prevent having the pidfile being set in two different locations, it would be nice if the osqueryd application settings could be stored with the standard key=value approach. This would allow the init script to source the config at runtime. The downside however would that the init script still makes the assumption that the location of osquery.conf is standardized.
I'm not really sure why the pidfile needs to go in the osquery directory anyway, considering most pidfiles for daemons like this typically exists in /var/run/$program.pid which actually would be a nicer default as it removes the requirement of having the folder there in /var. I'd prefer to not keep osquery in /var anyway.
same goes for the lockfile. Typically you'd never bother with this setting and you'd keep it in /var/lock/subsys/$progname
