FreeBSD tuning(7) is largely out of date

[Joe Greco]

I've been spending some time trying to determine some tuning issues w.r.t.
moving to FreeBSD 5.*, and I've run into some frustrations.

Perhaps the biggest one is that there is a severe lack of information on
how to tune large systems in tuning(7) and/or the Handbook, parts of which
seem to be direct from tuning(7).

I've read numerous mailing list posts which say things like "See the
tuning(7) man page and innumerable mailing lists posts" but the reality is
that most of the mailing list posts do not serve as reasonable
documentation.

Let's take a small example of something like KVA_PAGES.  Under 4.11, I read
the "innumerable mailing lists posts" which suggested that the default of
256 could be bumped up to 384 or 512.  Being conservative, I went to 384.
The machine wouldn't boot.  After much searching, I located *in* *the*
*source* that if you're using PAE, KVA_PAGES needs to be multiplied by two.
768 got me what I needed.

So, that brings me here:

Is there someone around who is actually familiar with large system tuning
and who might be willing to update the tuning info with a comprehensive set
of items that are likely to need tweaking on larger servers?  I'm not
talking about "let's write a book."  I'm talking more like "edit
/sys/i386/conf/NOTES, fix KVA_PAGES to mention the PAE caveat, maybe
elaborate just a tad more, and then stick a terse reference such as 
'Changing the size of the kernel's virtual address space can be done via 
KVA_PAGES.  See /sys/i386/conf/NOTES' into tuning(7)".  Repeat for as 
many tunables as possible.  Maybe add a few sentences about any common
caveats or gotchas.  I would just like to see the tunables all mentioned
in tuning(7).  A short paragraph about each would be heaven.

I don't mind paying a qualified someone to do this if they can do a good 
job of it at a reasonable rate.  Contact me off-list if interested.

Thanks.

... JG
-- 
Joe Greco - sol.net Network Services - Milwaukee, WI - http://www.sol.net
"We call it the 'one bite at the apple' rule. Give me one chance [and] then I
won't contact you again." - Direct Marketing Ass'n position on e-mail spam(CNN)
With 24 million small businesses in the US alone, that's way too many apples.