Dave Ewart
2014-07-04 08:44:35 UTC
Hello,
I've been getting repeated diffs in my Rancid reports because our
switches use dynamic VLANs. Bascially, the output from 'show vlan' is
constantly varying and there is no need to capture it: its results are
simply noise, for our purposes.
So I've been wondering about the *correct* way to prevent that output
hitting the diffs.
I've done the following:
- Copied /usr/lib/rancid/bin/rancid to a new file named
/usr/lib/rancid/bin/rancid-custom
- Modified router.db so that it refers to 'cisco-custom' instead of
'cisco'
- Modified /usr/lib/rancid/bin/rancid-fe to add an extra link from
'cisco-custom' to 'rancid-custom'
- Then edit /usr/lib/rancid/bin/rancid-custom to remove the 'show vlan'
output.
This works and does what I want, so far so good.
This seems like a robust solution, but because I've modified stock
files, I'm going to miss out on any updates to the rancid installation
that the Debian packages bring, next time there's an update.
Is my approach reasonable or can you suggest a better, more robust
approach?
Cheers,
Dave.
I've been getting repeated diffs in my Rancid reports because our
switches use dynamic VLANs. Bascially, the output from 'show vlan' is
constantly varying and there is no need to capture it: its results are
simply noise, for our purposes.
So I've been wondering about the *correct* way to prevent that output
hitting the diffs.
I've done the following:
- Copied /usr/lib/rancid/bin/rancid to a new file named
/usr/lib/rancid/bin/rancid-custom
- Modified router.db so that it refers to 'cisco-custom' instead of
'cisco'
- Modified /usr/lib/rancid/bin/rancid-fe to add an extra link from
'cisco-custom' to 'rancid-custom'
- Then edit /usr/lib/rancid/bin/rancid-custom to remove the 'show vlan'
output.
This works and does what I want, so far so good.
This seems like a robust solution, but because I've modified stock
files, I'm going to miss out on any updates to the rancid installation
that the Debian packages bring, next time there's an update.
Is my approach reasonable or can you suggest a better, more robust
approach?
Cheers,
Dave.
--
Dave Ewart
***@ceu.ox.ac.uk
Computing Manager, Cancer Epidemiology Unit
University of Oxford
N 51.7516, W 1.2152
Dave Ewart
***@ceu.ox.ac.uk
Computing Manager, Cancer Epidemiology Unit
University of Oxford
N 51.7516, W 1.2152