Don't forget that TACACS can log all commands entered into a router. When
used in combination with rancid and cvs/cvs-web, it's very useful.
I'm looking for a simple tool, in which each and every one has to
manually record whatever (s)he has done or any incident (s)he observed
so that the tool archives that data someway. Later, in case if someone
needs, (s)he should be able to search for that archive by date, by
person, by a random phrase, etc.
Hmm; thanks (I never met rancid before). I wonder, how people can even work
without such tools... -
It (expect script, tftpd, cvs web) was fone of the first things which I
implemented, when returned backk to the network administration...
The only problem is cvsweb, which can not work remotely (it uses rcs instead
of cvs).
both of which can be sufficiently tweaked to meet your device audit needs.
(A SourceForge loyalist, but I'm a RANCID kind of girl, myself...)
And, of course, remember the least costly and most oft overlooked practice
of establishing solid policies. Tools should be deployed to enforce a
well-defined policy, including guidelines and procedures laying down the law
when it comes to change management and change control of production devices.
You mentioned an outlet for _manual_ recording/documentation of laying on
of hands befalling the nodes, so define a must-have and must-do list
holding dominion over such activity, requiring that appropriate backups
occur, backouts are ready to go when things burst into flames, and that
all work be delineated and documented explicitly ex post facto.
Then, sit back and enjoy the grumbling of your paperwork-hating
associates, and be prepared to crack skulls if they flake on updating the
change control machanisms, as set forth in the unbudging monolith that is
your change management policy.
Still liking TACACS-RANCID though, as you can lead a horse to water, but
you can't make him think...
--ra
On Tue, Nov 25, 2003 at 03:54:34PM -0700, guy said something to the effect of:
