RANCID is used to make backups of switch and router configs, and notify admins of changes to those configs every 30 minutes.
Service directory is /afs/bx.psu.edu/service/rancid
There is a v/ directory with different versions. prod/ is a symlink to v/<something>
The etc and var directories are mountpoints for svc.rancid.etc and svc.rancid.var
There is a kerberos principal and LDAP entry for rancid-login which lets rancid log in to switches configured for RADIUS. The username/password for rancid-login and for switches not configured to do RADIUS is stored in /afs/bx.psu.edu/service/rancid/.cloginrc
cfengine configures the cron job under app_rancid to run every 30 minutes. This cron job requires a keytab, and uses k5start to call /afs/bx.psu.edu/service/rancid/prod/etc/run.sh. The run.sh script sets up the environment, and also rotates the log files daily.
We're using a version of RANCID with git extensions (https://github.com/dotwaffle/rancid-git). It is left as an exercise to the reader to figure out how to make use of that. The configure script is called something like:
And then installed with DESTDIR=something make install and copied to its final destination, as the svc.rancid volume is RO replicated.
There are also dlogin and drancid scripts that were added afterwards to support the Dell PowerConnect 6248. These can just be copied around when upgrading. You also need to add the 'dell' vendor to rancid-fe.