pf side effects

From: Anthony Ginepro <anthony.ginepro_at_laposte.net>
Date: Tue, 4 May 2004 19:32:08 +0200
I'm not experienced enough to patch these problems so I share my experiments
in the hope that hackers could have clues about stability with pf.

Since using pf on -CURRENT, I'm having trouble with bluetooth LAN. I can't 
directly translate ipf rules in pf rules because pf doesn't accept rules on 
interfaces that don't exist yet (tun0). It's certainly inherents to pf so I've
found another way around this.

However ppp core dumps now regularly when initiating a network connection from
the bluetooth device (palm) with this stack :

(gdb) where
#0  0x08063f38 in NgRecvData ()
#1  0xbfbfe284 in ?? ()
#2  0x0806a1a5 in NgRecvData ()
#3  0x0806324b in NgRecvData ()
#4  0x08063df8 in NgRecvData ()
#5  0x0806aa00 in NgRecvData ()
#6  0x0806f7b6 in NgRecvData ()
#7  0x0806f6c0 in NgRecvData ()
#8  0x08078366 in NgRecvData ()
#9  0x0805d2b7 in NgRecvData ()
#10 0x0804d896 in NgRecvData ()
#11 0x080722f5 in NgRecvData ()
#12 0x08071e93 in NgRecvData ()
#13 0x0804b77c in NgRecvData ()
I don't know what other information I could send to help people about this.

I had also oidentd not starting when using flag "-m", complaining about /dev/kmem
not present.

When using ipf, ppp & oidentd work correctly.

Anthony.
Received on Tue May 04 2004 - 08:32:11 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:37:53 UTC