Playing around a little with a lab network of virtual centos 5.3 machines.

I built a test DNS server and decided to see what the caching-nameserver.rpm package was all about.

And, of course, it works. Pretty painless if all you need is a caching server.

However, I'm a bit baffled how it is working. The only config file it creates is /var/named/chroot/etc/named.caching-nameserver.conf (with a sym link in /etc).

So I have two questions:

1. How does bind know to use this config file? I thought bind only looked for /etc/named.conf.

2. The symlink in /etc is actually to /var/named/chroot//etc/named.caching-nameserver.conf. Why the double slash after /var/named/chroot?

1-Actually it shouldn't if I'm not mistaken. Are you sure there isn't a named.conf anywhere?

2-No idea, I guess because is a chroot environment?

Cheers!

The caching-nameserver package is supposed to provide you with simple BIND (named) configuration so you won't have to write zone datafiles from scratch
(zone datafiles included in the package also have some default TTL values for DNS resource records, in case you don't have a clue what the right values should be).

As for the named.caching-nameserver.conf file is concerned, it there with pretty much the same idea as the zone datafiles provided with the package - to give you something to start with.
To my knowledge, 'named' always uses /etc/named.conf file (which is actually symbolic link to /var/named/chroot/etc/named.conf - if 'named-chroot' package is installed). So feel free to rename it (to named.conf) for use as named config file.

I have absolutely no idea what's with the double slash.

If you're interested in this subject, i recommend reading "DNS & BIND" - OReilly Media.


Regards,