Re: pf(4) and altq in -current

["John D. Baker" <jdbaker%consolidated.net@localhost>]

Did you ever find out anything about your problems with (then) -current
with pf(4) and your altq rules?

I've started testing 10.0_BETA on my router and the result so far is a
dismal failure.  At present I have my custom 10.0 kernel and modules
running with an up-to-date 9.3_STABLE userland (so I can return to the
working configuration with a simple reboot).

I'm not getting any error messages, warnings or notifications from the
kernel or pf subsystems, but a simple 'ping' test from my internal
network to a well-known external host averages over 4000ms RTT with 60%
to 70% packet loss.  The same test from the router itself is a steady 19ms
RTT with no packet loss.

I usually test -current (and X.0_BETA) builds by net-booting them and
have a "pf.boot.conf" file tailored to allow NFS to work through the
firewall.  The above problems are likely why it never finished booting
and looking at that machine's "/var/run/rc.log" on the file server showed
it was stuck running 'pf_boot'.

(While finishing the update of the NFS-resident system, I discovered that
I hadn't netbooted the router since the 8.99.xx era.)

-- 
|/"\ John D. Baker, KN5UKS               NetBSD     Darwin/MacOS X
|\ / jdbaker[snail]consolidated[flyspeck]net  OpenBSD            FreeBSD
| X  No HTML/proprietary data in email.   BSD just sits there and works!
|/ \ GPGkeyID:  D703 4A7E 479F 63F8 D3F4  BD99 9572 8F23 E4AD 1645