[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
kernel/2172: crash and freeze, maybe pf and/or ip6 related
- To: gnats@openbsd.org
- Subject: kernel/2172: crash and freeze, maybe pf and/or ip6 related
- From: clay@boobah.com
- Date: Wed, 7 Nov 2001 23:13:58 -0500 (EST)
- Resent-Date: Wed, 7 Nov 2001 21:30:02 -0700 (MST)
- Resent-From: gnats@cvs.openbsd.org (GNATS Management)
- Resent-Message-Id: <200111080430.fA84U2NQ022109@cvs.openbsd.org>
- Resent-Reply-To: gnats@cvs.openbsd.org, clay@boobah.com
- Resent-To: bugs@cvs.openbsd.org
>Number: 2172
>Category: kernel
>Synopsis: crash and freeze, maybe pf and/or ip6 related
>Confidential: no
>Severity: serious
>Priority: medium
>Responsible: bugs
>State: open
>Class: sw-bug
>Submitter-Id: net
>Arrival-Date: Wed Nov 7 21:30:01 MST 2001
>Last-Modified:
>Originator: Clay Caviness
>Organization:
net
>Release: -current (nov 4)
>Environment:
System : OpenBSD 3.0
Architecture: OpenBSD.i386
Machine : i386
>Description:
my machine is freezing every 2 days or so (this is the third crash in the past week)
with the message "panic: m_copydata: off -1 < 0"
looks to my extremely untrained eye to be pf and/or ipv6 related. i have an ipv6
tunnel provided by he.net, but haven't added any pf rules specific to it.
i have the crash dump saved, so i can do any other diagnostics needed.
here is the trace:
(gdb) target kcore bsd.0.core
panic: m_copydata: off -1 < 0
#0 0x1000 in ?? ()
(gdb) where
#0 0x1000 in ?? ()
#1 0xe02d1db9 in boot ()
#2 0xe017ee07 in db_boot_dump_cmd ()
#3 0xe017e9b0 in db_command ()
#4 0xe017ec33 in db_command_loop ()
#5 0xe0181eb6 in db_trap ()
#6 0xe02ccf62 in kdb_trap ()
#7 0xe02dab24 in trap ()
#8 0xe0100db9 in alltraps ()
#9 0xe01b18c9 in panic ()
#10 0xe01c1c80 in m_copydata ()
#11 0xe0176d9a in pf_pull_hdr ()
#12 0xe0176766 in pf_test_state_icmp ()
#13 0xe017783f in pf_test6 ()
#14 0xe02b6b12 in ip6_input ()
#15 0xe02b69af in ip6intr ()
#16 0xe010203e in Xsoftnet ()
(gdb)
here is ps (it's long)
{clay@scootch}[/var/crash]$ sudo ps kaux -M bsd.0.core
USER PID %CPU %MEM VSZ RSS TT STAT STARTED TIME COMMAND
cyrus 23946 0.0 0.0 4440 0 ?? Is 6:30PM 0:00.26 (imapd)
root 0 0.0 0.0 0 0 ?? DKLs Mon09AM 0:00.07 (swapper)
root 1 0.0 0.0 332 0 ?? Is Mon09AM 0:00.10 (init)
root 2 0.0 0.0 0 0 ?? DKL Mon09AM 0:00.01 (pagedaemon)
root 3 0.0 0.0 0 0 ?? DKL Mon09AM 0:00.70 (reaper)
root 4 0.0 0.0 0 0 ?? DKL Mon09AM 0:00.04 (cleaner)
root 5 0.0 0.0 0 0 ?? RKL Mon09AM 0:08.39 (update)
root 6 0.0 0.0 0 0 ?? DKL Mon09AM 0:00.01 (crypto)
root 7 0.0 0.0 0 0 ?? RKL Mon09AM 0:01.04 (apm0)
root 8 0.0 0.0 0 0 ?? DKL Mon09AM 0:00.01 (usb0)
root 22272 0.0 0.0 100 0 ?? Rs Mon09AM 0:01.18 (syslogd)
root 7096 0.0 0.0 328 0 ?? Rs Mon09AM 0:05.25 (pflogd)
named 5807 0.0 0.0 1060 0 ?? Rs Mon09AM 0:08.04 (named)
root 10238 0.0 0.0 64 0 ?? Is Mon09AM 0:00.00 (portmap)
root 22681 0.0 0.0 356 0 ?? Is Mon09AM 0:00.01 (mountd)
root 7005 0.0 0.0 124 0 ?? Is Mon09AM 0:00.00 (nfsd)
root 1338 0.0 0.0 108 0 ?? IL Mon09AM 0:00.00 (nfsd)
root 14368 0.0 0.0 108 0 ?? IL Mon09AM 0:29.95 (nfsd)
root 3197 0.0 0.0 108 0 ?? IL Mon09AM 0:00.00 (nfsd)
root 32152 0.0 0.0 108 0 ?? IL Mon09AM 0:00.14 (nfsd)
root 19352 0.0 0.0 232 0 ?? Ss Mon09AM 0:00.22 (dhcpd)
root 24511 0.0 0.0 1356 0 ?? Rs Mon09AM 0:04.40 (httpd)
root 5694 0.0 0.0 80 0 ?? Is Mon09AM 0:00.31 (inetd)
root 13281 0.0 0.0 332 0 ?? Is Mon09AM 0:02.89 (sshd)
root 23220 0.0 0.0 636 0 ?? Ss Mon09AM 0:05.57 (ntpd)
root 4305 0.0 0.0 388 0 C0- I Mon09AM 0:00.02 (sh)
mysql 15709 0.0 0.0 11568 0 C0- I Mon09AM 0:12.81 (mysqld)
root 22738 0.0 0.0 4076 0 ?? Rs Mon09AM 3:23.23 (snort)
root 23490 0.0 0.0 1728 0 ?? Is Mon09AM 0:00.00 (smbd)
root 10048 0.0 0.0 872 0 ?? Rs Mon09AM 0:28.03 (nmbd)
root 3848 0.0 0.0 224 0 ?? Is Mon09AM 0:00.84 (cron)
root 11204 0.0 0.0 44 0 C0 Is+ Mon09AM 0:00.01 (getty)
root 2426 0.0 0.0 44 0 C1 Is+ Mon09AM 0:00.00 (getty)
root 19651 0.0 0.0 44 0 C2 Is+ Mon09AM 0:00.01 (getty)
root 30626 0.0 0.0 44 0 C3 Is+ Mon09AM 0:00.00 (getty)
root 17516 0.0 0.0 44 0 C5 Is+ Mon09AM 0:00.00 (getty)
root 28667 0.0 0.0 496 0 ?? Is Mon09AM 0:06.75 (screen-3.9.8)
clay 19689 0.0 0.0 27468 0 p1 Ss+ Mon09AM 0:36.98 (tf)
clay 24035 0.0 0.0 388 0 p2 Is Mon09AM 0:00.16 (ksh)
clay 7463 0.0 0.0 388 0 p3 Is+ Mon09AM 0:00.18 (ksh)
root 208 0.0 0.0 916 0 ?? Ss Mon09AM 0:11.00 (sendmail)
cyrus 6186 0.0 0.0 832 0 ?? Is 10:06AM 0:01.70 (imapd)
root 1683 0.0 0.0 376 0 ?? I 10:08AM 0:03.34 (sshd)
clay 28649 0.0 0.0 376 0 p0 Is 10:08AM 0:00.01 (ksh)
clay 5389 0.0 0.0 288 0 p0 S+ 10:08AM 0:00.10 (screen-3.9.8)
www 22249 0.0 0.0 3620 0 ?? I 11:15AM 0:05.64 (httpd)
www 31216 0.0 0.0 3836 0 ?? I 11:16AM 0:06.26 (httpd)
www 22036 0.0 0.0 3796 0 ?? I 11:21AM 0:04.28 (httpd)
www 9606 0.0 0.0 3412 0 ?? I 11:26AM 0:03.65 (httpd)
www 18536 0.0 0.0 3748 0 ?? I 11:27AM 0:05.61 (httpd)
www 24694 0.0 0.0 3848 0 ?? I 12:58PM 0:03.21 (httpd)
www 3576 0.0 0.0 3492 0 ?? I 1:21PM 0:02.18 (httpd)
clay 9586 0.0 0.0 68 0 p2 I+ 3:52PM 0:00.01 (man)
clay 6837 0.0 0.0 372 0 p2 I+ 3:52PM 0:00.00 (sh)
clay 5677 0.0 0.0 152 0 p2 I+ 3:52PM 0:00.04 (more)
root 1108 0.0 0.0 240 0 ?? Z - 0:00.00 (cron)
here is my (very basic) pf.conf:
{clay@scootch}[/var/crash]$ cat /etc/pf.conf
# $OpenBSD: pf.conf,v 1.2 2001/06/26 22:58:31 smart Exp $
#
# See pf.conf(5) for syntax and examples
ext_if = "fxp0"
int_if = "fxp1"
ext_ip = "X.X.X.X/32"
int_ip = "10.0.0.0/8"
spoofed = "{ 10.0.0.0/8, 172.16.0.0/12, 192.168.0.0/16, \
255.255.255.255/32 }"
# normalize all incoming traffic
scrub in on $ext_if all
# block by default
block in all
# block non-routable packets coming in from the mean world
block in quick on $ext_if from $spoofed to any
# block smb, nfs, mysql, rndc? from the mean world
block in quick on $ext_if proto tcp from any to any port {137, 138, 139, 901}
block in quick on $ext_if proto tcp from any to any port {2049, 111}
block in quick on $ext_if proto tcp from any to any port 3306
block in quick on $ext_if proto tcp from any to any port 953
pass out on $ext_if proto tcp all modulate state
pass out on $ext_if proto udp all keep state
# pass all packets in and out (these are the implicit last two rules)
pass in all
pass out all
>How-To-Repeat:
turn on machine and wait - i've never been active when it's crashed, so i haven't
seen anything repeatable.
>Fix:
sorry, can't help you here
>Audit-Trail:
>Unformatted: