Configuring Rancid

In my last post I talked about Installing Rancid on Centos 7 so be sure to check that out. In this post we now want to actually use Rancid so I’ll walk through on how to setup Rancid to login into network equipment and get configurations.

To recap: Rancid is a great tool to help monitor a device configuration for any changes. It also keeps track of them by using CVS (Concurrent Version System) for backups, so you can go back and compare versions or revert to a previous configuration. Rancid supports multiple hardware from Cisco, HP, Dell, Juniper and more. This is all open-source so you can create custom scripts or add commands to really make this a personal repository that fits your company.

Rancid has a couple of areas that need to be edited before we can let it run. The first area is the router.db file. This file is in each “group” or folder that rancid stores. In our example we have two groups’ routers and switches. This file has all the devices you want rancid to keep track of. We must edit this file with either the DNS name of the device or IP address. To do that go to the following:

Login into Rancid with the Root account followed by su – rancid Type of Device is the “group” you configured the List of Groups under the rancid.conf file during the installation

nano /usr/local/rancid/var/{Type of Device}/router.db

At first this is a blank file so nothing will be in the file, below is an example of what I have done in the past

# This is an example of a router.db fle in Rancid.
 
# Location of File /usr/local/rancid/var/Routers/router.db
 
# Type the Router name or IP followed by the type of device. (Rancid supports more than just Cisco)
 
# Personally I think it looks better with DNS names.
 
Router01.example.com;cisco;up;
 
# You can also mark a device as down. This will cause Rancid to skip it.
 
Router02.example.com;cisco;down;
 
Router03.example.com;cisco;up;
 
# Of course you can also comments throughout this file or alongside devices for notes.
 
# Edge Switches for Computers at HQ
 
Switch-HQ-Edge14.example.com;cisco;up; LOCATION: Server Rack A
 
Switch-HQ-Edge07.example.com;cisco;up; LOCATION: Room 101

Switch-HQ-Edge21.example.com;cisco;down; LOCATION: Building A (Under Construction)

Once you add the all the devices in each Router.db we now need to look at the .clogin file which is located (/home/rancid/.cloginrc) this file will store the usernames and passwords for each device,  this file supports wildcards so if you have a good naming scheme use it! I have had luck with commenting out everything in that file and just starting at the bottom of the file with my custom configuration, such as:

# Custom Configuration at the bottom of .cloginrc
 
# No Enable Prompts
 
add noenable Swich-E-Edge05.example.com 1
 
# Connect to All CORES (Notice the wildcard, remember you still need to add the full name to the router.db file.)
 
add user S-core0*.example.com username-goes-here
 
add password S-core0*.example.com password-goes-here
 
add method S-core0*.example.com ssh
 
add noenable S-core0*.example.com 1
 
# Connect To All Campus-5-Edge Switches
 
add user Switch-C5-Edge*.example.com username-goes-here
 
add password Switch-C5-Edge*.example.com password-goes-here enable-password-here
 
add method Switch-C5-Edge*.example.com ssh

So we now have everything ready to test run rancid, we can test our luck by just running it once and to do that run the following command below. When the commands completes you should be able to browse via the webpage, see the devices you added with their configurations, and get an email if you’ve followed by earlier post.

/usr/local/rancid/bin/rancid-run

For reference rancid keeps a log each time it runs and each different of group or device. To view the log use my favorite Linux test editor nano. :) Or you can cat it!

nano /usr/local/rancid/var/logs/{Type of Device/Group}

In this example log file everything went well, rancid is pretty good at telling you what the issue is if there is one.

starting: Sun Feb 15 18:01:56 MST 2015
 
Trying to get all of the configs.
 
All routers sucessfully completed.
 
cvs diff: Diffing .
 
cvs diff: Diffing configs
 
cvs commit: Examining .
 
cvs commit: Examining configs
 
ending: Sun Feb 15 18:02:45 MST 2015

Once you capture devices you can view their configuration on the browser by going to http(s)://rancid-server-ip/viewvc. In this example I just have one device with two revisions on it, I’m using router03.example.com. What do you think?

Screenshot of Rancid in ViewVC, using Router03 as an example.

Diff between revisions on Router03.

One of the last things to do is to setup a cron job to run rancid, it depends on your environment but in this example I’m running the rancid job every hour on the hour. I’m also clearing the log files every five days but running the job every day day at 11:50pm. Below is my example of the cron job. Make sure you in the rancid account (su - rancid) and then type crontab -e. Note you will be using vi so here is a crash course once you in the file type “i” for INSERT MODE type out the configuration once you are done type Escape key followed by “:wq” which will write (save) and quit the vi program.

#
# Rancid User Crontab File
#
 
# Begin Hourly Router Dump
1 * * * * /usr/local/rancid/bin/rancid-run
 
# Begin Cleanup of Differ Logs Every 5 Days
50 23 * * * /usr/bin/find /usr/local/rancid/var/logs -type f -mtime +5 -exec rm -rf {} \;

We now have a system that will log in and grab configurations of our network equipment on a scheduled basis. Rancid does way more than just Cisco devices although its in the name :). Rancid has made it pretty easy to even add you own custom scripts or add-on to what you want to see when rancid pulls a configuration so like before check out Shrubbery at http://www.shrubbery.net/rancid/ and the threads from Rancid Discussions at http://www.gossamer-threads.com/lists/rancid/ for details. I’ll stop it here I hope this information is helpful and as always let me know if you have any questions.