Two problems with netbsd-5

[Dominik BiaÅy <dmb%yenn.underlegend.net@localhost>]

I'm getting panic, I have this output from gdb:

yenn# gdb netbsd.0
GNU gdb 6.5
Copyright (C) 2006 Free Software Foundation, Inc.
GDB is free software, covered by the GNU General Public License, and you are
welcome to change it and/or distribute copies of it under certain conditions.
Type "show copying" to see the conditions.
There is absolutely no warranty for GDB.  Type "show warranty" for details.
This GDB was configured as "i386--netbsdelf"...
(gdb) target kvm netbsd.0.core
#0  0xc057272f in cpu_reboot ()
(gdb) bt
#0  0xc057272f in cpu_reboot ()
#1  0xc04c65ac in panic ()
#2  0xc0575398 in trap ()
#3  0xc010cb1f in calltrap ()
#4  0xc056da41 in db_read_bytes ()
#5  0xc01eb7da in db_get_value ()
#6  0xc056e2ca in db_stack_trace_print ()
#7  0xc04c6581 in panic ()
#8  0xc0575398 in trap ()
#9  0xc010cb1f in calltrap ()
#10 0xc04e9815 in m_tag_find ()
#11 0xc01bc051 in ipsec_delaux ()
#12 0xc01a526b in nd6_output ()
#13 0xc0172999 in fr_fastroute ()
#14 0xc0172fb9 in fr_send_ip ()
#15 0xc01737bf in fr_send_reset ()
#16 0xc01708a1 in fr_check ()
#17 0xc0172200 in fr_check_wrapper6 ()
#18 0xc053f951 in pfil_run_hooks ()
#19 0xc019b69e in ip6_input ()
#20 0xc019c41a in ip6intr ()
#21 0xc04aa91c in softint_dispatch ()
#22 0xc0100e0d in Xsoftintr ()
#23 0x00000000 in ?? ()
(gdb)

I see there code for sending rst from ipfilter, no? It has started to
happen after 54 days of uptime, mabe a day after I touched firewall to
send rst on blocking on some port. The firewall has been sending rst
already, however, for all blocked ports beyond 1024. Maybe just
nothing has been triggering it until yesterday. Of course I'm mistaken
most probably.

Ah! This is only ipf6.conf! ipf.conf works fine, I have sending rst on
block there everywhere in the file.

After reboot, the panic showed again after a few hours.

This was NetBSD 5.0_BETA from around 2 November. Now it runs kernel
from yesterday and doesn't panic (yet). The kernel is a bit modified
GENERIC. This is on i386.

Second problem. altqd doesn't shutdown. It just hangs forever, I need
to pkill -9 altqd, and then /etc/rc.d/altqd start -- to reload altqd.
In /etc/rc.shutdown I need to have pkill -9 altqd, or the system won't
shutdown. I had this problem in netbsd-4, too.

My altq.conf:

interface ex0 bandwidth 512K tbrsize 8K hfsc
class hfsc ex0 def root pshare 25 red default
class hfsc ex0 srv root pshare 25 red
        filter ex0 srv 0 0 XX.XX.XX.XX 0 0 dontwarn
class hfsc ex0 fermi root pshare 25 red cleardscp
        filter ex0 fermi 0 0 XX.XX.XX.XX 0 0 tos 0x84 tosmask 0x84 dontwarn
class hfsc ex0 planck root pshare 25 red cleardscp
        filter ex0 planck 0 0 XX.XX.XX.XX 0 0 tos 0x88 tosmask 0x88 dontwarn
        # eth openvpn
        filter ex0 planck 0 0 XX.XX.XX.XX 1194 0 dontwarn
conditioner ex0 cdnr <tbmeter 1M 8K <pass><drop>>
        filter ex0 cdnr 0 0 0 0 0 dontwarn

interface ex1
conditioner ex1 mark_unknown <mark 0x44>
        filter ex1 mark_unknown 0 0 192.168.0.0 netmask 0xffff0000 0 0 dontwarn
conditioner ex1 mark_fermi <mark 0x84>
        filter ex1 mark_fermi 0 0 192.168.0.2 0 0 dontwarn
conditioner ex1 mark_planck <mark 0x88>
        filter ex1 mark_planck 0 0 192.168.0.3 0 0 dontwarn
        filter ex1 mark_planck 0 0 192.168.0.4 0 0 dontwarn

The dmesg is at http://dominik.bialy.info/dmesg

Sorry for my fatal English.

I write to NetBSD lists seldom, so using occasion, I'd like to whish
happy New Year to all NetBSD hackers and users around the Globe :)

Thanks
        Dominik
-- 
Dominik Biały . dmb%yenn.underlegend.net@localhost . dmb@IRCnet
DMB-RIPE      . dominik%bialy.info@localhost       . gg:8194865