[prev in list] [next in list] [prev in thread] [next in thread] 

List:       openbsd-sparc
Subject:    Re: bridge changes traffic interface for pf, but not for tcpdump
From:       Jim Fron <j-fron.q.public () comcast ! net>
Date:       2005-07-06 22:43:29
Message-ID: 69C9801A-E031-4994-A583-3A48E960992A () comcast ! net
[Download RAW message or body]

On Jul 6, 2005, at 9:03 AM, Todd C. Miller wrote:

> In message <20a9f61fad575ba5f95b3ab9a57a0c54@comcast.net>
>     so spake Jim Fron (j-fron.q.public):
>
>
>> Yes, I'm getting the feeling that what I'm seeing is "not  
>> normal."  As
>> I've said, I have a suspicion that it's due to the le[dma] SBUS
>> interfaces not having their own MAC address, and that somehow getting
>> confused at the bridge level.  I'm thinking about getting a QFE to  
>> test
>> this out.
>>
>
> Have you tried setting local-mac-address?=true in the eeprom?
>
>  - todd

Nothing quite so simple as that, really... it doesn't apply.  The old  
le/ledma SBUS interfaces don't have their own mac-address; setting  
local-mac-address? to true has no effect when the controller has no  
mac-address of its own -- all three still get the SparcStation's on- 
board address.  Hence my thought about getting a QFE to resolve the  
issue.  Haven't gotten around to that, though.

To head off further questions:

1. IIRC, No, changing the mac-address in the running system before  
the secure level is set only makes things look ookie: the OBSD  
machine's `tcpdump` reports the "new" hardware address, but it is  
"actually" sent on the network as the machine's "default" mac- 
address, since the transciever supplies the only address it knows of  
(tested by plugging in interface directly to a hub connected to only  
one other machine -- no network loops).

2. I believe I solved all of this with a kernel patch that no one  
particularly liked, but which resolved the problem for me.  I expect  
the problem will ultimately be solved when all of SBUS-based Sparc  
hardware in the world fails due to old age.  In the mean time, I seem  
to be one of about two or three people who care, and I have a  
solution, so I don't much care any more.

Thanks, though.  I appreciate the call-back to an old thread.
JMF

[prev in list] [next in list] [prev in thread] [next in thread]