direction for training

Matthew Seaman m.seaman at
Sun Jul 4 08:35:18 PDT 2004

On Sun, Jul 04, 2004 at 09:31:03AM -0400, Dru wrote:

> Also, I see all 3 BSDs as BSD and would like to see a curriculum that 
> addresses all aspects of BSD. Do others agree that this shouldn't be a 
> FreeBSD-baby and should be open to input from all of the BSDs?

How much of what you need to know to be a FreeBSD admin is
specifically FreeBSD related?  A great deal of the material is not
just common to all the *BSD family, but to any Unixoid system (shell
scripting for example) or to any sort of computer.

Consider, for example vinum(8).  To use vinum, you should understand
concepts like "what is RAID 1", "hardware vs software raid" and "what
are the pros- and cons- of using volumne management", none of which
are FreeBSD specific.  Background knowledge should include what
alternative RAID systems are available on various different OSes -- to
just list a few of the software RAID systems: Solaris ODS, Vertias
VxVM, vinum(8), RaidFrame, LVM, even if it doesn't go much beyond a
one-liner describing each of them.

Such is the basic theory that anyone working on any sort of system
would have to learn about first.  *BSD specific material then follows
in, say, the practical details of using vinum -- how to build a RAID
filesystem, installing the system using a mirrored root, recovery from
a failed hard drive, interaction with GEOM under 5.x, etc.  That's not
to mention related things like ccd(4) or atacontrol(8).

Such knowledge is purely factual, and relatively easily to test
on-line.  Similar modules could be constructed for other areas of
system function, and presumably students could work towards a diploma
or somesuch by getting a passing grade in a certain number of modules.

However, at some point it will become more valuable to the student or
to any prospective employer to add modules covering implementations on
many different OSes.

Going beyond mere regurgitation of facts and starting to do useful
problem solving requires integration of knowledge from many such
modules.  A good sysadmin has to be able to create a synthesis of
their general knowledge on a wide range of subjects, and the specific
information about hardware and software they know or can gather
through research in order to solve problems like

    "How can we increase the IO throughput on our database service
     with minimal cost and minimal disruption to live services"


    "spec. out the hardware and software needed to provide an e-mail
     service for a company of such and such size, that sends so many
     messages per day, allowing for a reasonable expected growth over
     the next 18 months"

This third level of competence is where the real value of the
qualifications would lie.  However it's the hardest level to assess
anyone at -- questions may not have obviously right or wrong answers,
and can't be tested using multiple choice.  It's also impossible to do
well without practical experience, either in a lab setting or simply
by learning on the job.



Dr Matthew J Seaman MA, D.Phil.                       26 The Paddocks
                                                      Savill Way
PGP:         Marlow
Tel: +44 1628 476614                                  Bucks., SL7 1TH UK
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 187 bytes
Desc: not available
Url :

More information about the freebsd-advocacy mailing list