Subject: Re: IPSEC still fails on BETA2/vax
To: None <itojun@iijlab.net>
From: Olaf Seibert <rhialto@polderland.nl>
List: current-users
Date: 07/11/2002 00:08:37
On Wed 10 Jul 2002 at 08:16:10 +0900, itojun@iijlab.net wrote:
> try raising net.key.larval_lifetime to 120 (or 300?) and see if it
> makes a difference.
Yes, it does make a difference: the keys seem to get set now and the
crash I was experiencing happens now without explcit "setkey" command
but when I start pinging.
2002-07-10 23:58:24: DEBUG: sockmisc.c:570:sendfromto(): 1 times of 60 bytes message will be sent to 10.0.0.7[500]
2002-07-10 23:58:24: DEBUG: plog.c:193:plogdump():
a4cebac3 85bf7c77 3e4b367a 87122373 08102001 4d1e231e 0000003c c53d4e85
ff651d28 ec4b6d5c ba7ed809 1a1a812d d09480eb 18c311be d03479e6
2002-07-10 23:58:25: NOTIFY: isakmp.c:252:isakmp_handler(): the packet is retransmitted by 10.0.0.5[500].
bash-2.04# setkey -D
10.0.0.7 10.0.0.5
esp mode=transport spi=151932733(0x090e4f3d) reqid=0(0x00000000)
E: 3des-cbc 34f38daf a46a695b 8311c261 7e4f95f6 5f8bf96e 45f04ded
A: hmac-sha1 5ab502e8 5746bb64 97e90a99 b090214b dccf1b1e
seq=0x00000000 replay=4 flags=0x00000000 state=mature
created: Jul 10 23:58:20 2002 current: Jul 10 23:58:56 2002
diff: 36(s) hard: 43200(s) soft: 34560(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=1 pid=169 refcnt=1
10.0.0.5 10.0.0.7
esp mode=transport spi=108646114(0x0679cee2) reqid=0(0x00000000)
E: 3des-cbc 8aff8cbd a77277fd acf9b950 a94954b8 113c1df5 95e7eb65
A: hmac-sha1 f891ded0 f38914ab 674c6972 4e648c72 f3a6aaa1
seq=0x00000000 replay=4 flags=0x00000000 state=mature
created: Jul 10 23:58:19 2002 current: Jul 10 23:58:56 2002
diff: 37(s) hard: 43200(s) soft: 34560(s)
last: hard: 0(s) soft: 0(s)
current: 0(bytes) hard: 0(bytes) soft: 0(bytes)
allocated: 0 hard: 0 soft: 0
sadb_seq=0 pid=169 refcnt=1
bash-2.04# ping azenomei
p
85 RESTART SYS
p84 FAIL
83 BOOT SYS
-DKB0
>> NetBSD/vax boot [1.11 Sat Jun 15 18:30:21 UTC 2002] <<
tcpdump as seen from azenomei:
23:59:13.535207 xzan.falu.nl > azenomei.falu.nl: ESP(spi=151932733,seq=0x1)
23:59:13.536184 azenomei.falu.nl > xzan.falu.nl: ESP(spi=108646114,seq=0x1)
so it seems that it is the reception of the ESP packet which crashes
xzan (not the sending because it has been doing that a lot already in
the past without crashing).
I suppose I will add printf()s to the IPsec input path.
> itojun
-Olaf.
--
___ Olaf 'Rhialto' Seibert - rhialto@ -- Woe betide the one who feels
\X/ polderland.nl -- remorse without sin - Tom Poes, "Het boze oog", 4444.