Same boat.  $job has lots and lots of junk in %post to make machines
smell like us post-kick.  Bizarrely, we have been a cf2 shop for
nearly a decade.  The crux of this practice is that we kick machines
and bop their network off post-firstboot since they have 1.) a
'consider-it-compromised' root passwd (even though no passwd ssh is
allowed ... also configured @ %post) and 2.) no tripwire policy
instantiated with keys+passphrases.  Steps 1 and 2 of bringing our
machine to life post-kick is to set root passwd and bootstrap tripwire
interactively via iLO/DRAC (turn the network on before both of those
and youll get your knuckles rapped.  :)

Im a firm believer in disposable machines (no userland data on
direct-attached disk + 100% of running config in cfg mgmt + multiple
horizontal nodes == one can kick a node at any nearly anytime).  I
also think the several hundred lines of %post we have here is a bit
insane given _I_ find the function of ks is to slice disk and drop
enough binaries in-place to let cfg mgmt bootstrap and takeover the
rest.  Part of my selling $job to get us from here to there where we
dismantle all the %post cruft is getting the above three things in
place (data + runtime + multiples).  Once there, its sellable to $mgmt
that machines can have the kind of kicks you/I/whoever is talking
about in this thread.

FWIW ...

On Tue, May 4, 2010 at 5:45 PM, Matt Lawrence <[email protected]> wrote:
> I recently started a new job, getting information has been difficult and
> there is a bit of a turf war from the existing sysadmin, who is almost
> universally hated by the customer we both support.  So, this is a touchy
> situation politically.
>
> On the the issue.  I have finally gotten access to the kickstart files
> that are used to install most of the systems.  The one I am looking at
> right now is 1648 lines long, with about 1600 of them in the %post clause.
> I am of the opinion this is a bad idea, a kickstart shouldn't do much more
> than get a system up, running and able to talk to a configuration
> management system.  Naturally, there is no configuration management system
> and systems are left as initially installed for years.
>
> So, I'm looking for references to best practices that I can take to my
> boss and other management on the preferred way of doing RHEL kickstarts
> and configuration management.  Any suggestions?  TAL?
>
> -- Matt
> It's not what I know that counts.
> It's what I can remember in time to use.
> _______________________________________________
> Tech mailing list
> [email protected]
> http://lopsa.org/cgi-bin/mailman/listinfo/tech
> This list provided by the League of Professional System Administrators
>  http://lopsa.org/
>

_______________________________________________
Tech mailing list
[email protected]
http://lopsa.org/cgi-bin/mailman/listinfo/tech
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to