Skip site navigation (1)Skip section navigation (2)
Date:      Thu, 18 Aug 2005 17:38:57 +0200
From:      Peter van Dijk <peter@dataloss.nl>
To:        freebsd-current@freebsd.org
Subject:   'crash' with bridge.ko/tap; tcpdump problem
Message-ID:  <20050818153857.GB50070@dataloss.nl>

next in thread | raw e-mail | index | archive | help
Hi,

this message is quite uninformational; for that I apologize.

One of the reasons I upgraded my Sun Ultra5 to FreeBSD6 (BETA2) was
that I wanted working Ethernet bridging. Both ng_bridge and bridge.ko
cause nothing but instant panics on 5.4/sparc64.

The first thing I tried after upgrading to 6 was to load bridge.ko,
bridge tap0 and vlanX with it. When I typed 'ifconfig tap0' the box
froze as far as I could see. No monitor was attached at that time so
I'm afraid this is all the information I have.

I'm just posting this to note that not all of the bridge/ng_bridge
issues that 5.x had are resolved.

I should also note that tap-support in itself works fine with 6 (which
didn't with 5.4-RELEASE), and that if_bridge bridges my frames perfectly.

Another sidenote:
onion# tcpdump -nli bridge0 port 22
tcpdump: WARNING: bridge0: no IPv4 address assigned
tcpdump: verbose output suppressed, use -v or -vv for full protocol decode
listening on bridge0, link-type EN10MB (Ethernet), capture size 96 bytes
17:37:37.869099 [|ether]
17:37:38.776971 [|ether]
17:37:38.867750 [|ether]

I'm unsure whether this is if_bridge's fault or tcpdump's fault, but I've
heard from other people with this issue on x86 on real ethernet
interfaces. This tcpdump is from world, not ports.

Cheers, Peter
-- 
peter@dataloss.nl        | ~ tonight tonight, what is this potion
http://blog.dataloss.nl/ | ~ that makes a fool of me
UnderNet/#clue           |     Wayfinder, fr-025 soundtrack



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20050818153857.GB50070>