All of lore.kernel.org
 help / color / mirror / Atom feed
From: Matt Richards <matt@mattstone.net>
To: Andy Gospodarek <andy@greyhouse.net>, bridge@lists.osdl.org
Subject: Re: [Bridge] bridge wont forward when ip address changes
Date: Thu, 15 Feb 2007 03:55:06 +0000	[thread overview]
Message-ID: <45D3D99A.9050108@mattstone.net> (raw)
In-Reply-To: <bdfc5d6e0702131517v7e7ed5dbk8bb2f30d4d8842c0@mail.gmail.com>

[-- Attachment #1: Type: text/plain, Size: 2339 bytes --]

Andy Gospodarek wrote:
> On 2/10/07, Matt Richards <matt@mattstone.net> wrote:
>> hello :)
>>
>> I have been using the bridge utils for a long time now and its been very
>> good but there are a couple of issues that I came across when trying to
>> setup a wireless bridge...
>>
>> Firstly I currently have a setup like the following ...
>>
>> Computer A --- Wired Network --- linux box ---- wireless link ------
>> linux box ----- Wired Network --- Computer B
>>
>> I have bridged the wireless interfaces and the wired interfaces together
>> on both of the linux box's and data can travel across the wireless
>> bridge without any trouble but if I attempt to change the IP addresses /
>> subnet of Computers A and B the bridges seem to stop forwarding packets
>> and no data get through at all, its like they aren't learning the new
>> addresses.
>>
>> If I restart the 2 Linux boxes then everything starts flowing properly
>> again.
>>
>> The 'Linux boxes' are a couple of Motorola Wireless Access points
>> running OpenWRT.
>>
>> Does anybody know why this might be happening and what I might be able
>> to do about it?
>>
>
> I wonder if this is ARP related?  I've noticed some interesting
> behavior when bridging wired and wireless devices.  Have you noticed
> that you can communicate after the IP switch if you send a gratuitous
> ARP from any of the wireless devices?
>
hello,
no i haven't tried tried that, i did notice that if i try to make a
computer talk to another device after the ip change the arp table would
contain the IP with an invalid state and no mac address or a mac address
of all F's i think it was.

Sending arp packets wouldn't be a solution as this will be used in a
computer lab to connect sockets that are in parts of the room where its
not easy to run cables and therefore wouldn't be ideal for people to use
when learning about network operating systems.

One thing I have just thought of, while reading an email from the
openvpn mailing list, is that if i setup an openvpn tunnel with tap
interfaces over the wireless link and bridge the switches to the tap
interface then the problem should go away if it is to do with the
wireless interfaces.

Is a little weird tho, I had to lower to mtu to get the devices to ping
in the first place.

Matty.



[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

      parent reply	other threads:[~2007-02-15  3:55 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-02-11  4:19 [Bridge] bridge wont forward when ip address changes Matt Richards
     [not found] ` <bdfc5d6e0702131517v7e7ed5dbk8bb2f30d4d8842c0@mail.gmail.com>
2007-02-15  3:55   ` Matt Richards [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=45D3D99A.9050108@mattstone.net \
    --to=matt@mattstone.net \
    --cc=andy@greyhouse.net \
    --cc=bridge@lists.osdl.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.