[169532] in North American Network Operators' Group
Re: Managing IOS Configuration Snippets
daemon@ATHENA.MIT.EDU (Keegan Holley)
Fri Feb 28 22:20:42 2014
From: Keegan Holley <no.spam@comcast.net>
In-Reply-To: <20140301023520.GA3667@havarti.local>
Date: Fri, 28 Feb 2014 22:20:11 -0500
To: "Dale W. Carder" <dwcarder@wisc.edu>
Cc: nanog list <nanog@nanog.org>
Errors-To: nanog-bounces+nanog.discuss=bloom-picayune.mit.edu@nanog.org
On Feb 28, 2014, at 9:35 PM, Dale W. Carder <dwcarder@wisc.edu> wrote:
> Thus spake Keegan Holley (no.spam@comcast.net) on Fri, Feb 28, 2014 at =
09:49:19AM -0500:
>> I wasn=92t saying just fix it. I was saying that router configs =
don=92t lend well to versioning. =20
>=20
> Um, what?
>=20
> $> rlog r-cssc-b280c-1-core.conf | grep 'total revision'
> total revisions: 2009; selected revisions: 2009
I wish you were here to see my eyes rolling.. 2009 versions of something =
are no more grok-able than one current version. Congrats, you have a =
config backup system.
>=20
>> When it=92s a router config chances are someone fat-fingered =
something. Most of the time the best thing to do is to fix or at least =
alert on the error, not to record it as a valid config version.=20
>=20
> We have our operators manually check in revisions (think in rcs terms:
> co -l router, go do work, verify it, ci -u router) rather than
> unsolicited / cron-triggered checkins. Then the check-in message
> contains the operator's description text of the change and often a
> ticket number. So there slightly fewer fat-finger configs checked in.
That=92s not what the OP was looking for AFAIK. This is just change =
management.
>=20
> Dale
