Date:	Wed, 27 May 2009 10:25:14 +0200
From:	Jiri Pirko <jpirko@...hat.com>
To:	Andy Gospodarek <andy@...yhouse.net>
Cc:	netdev@...r.kernel.org, jgarzik@...ox.com, davem@...emloft.net,
	shemminger@...ux-foundation.org, bridge@...ts.linux-foundation.org,
	fubar@...ibm.com, bonding-devel@...ts.sourceforge.net,
	kaber@...sh.net, mschmidt@...hat.com, dada1@...mosbay.com
Subject: Re: [PATCH net-next] bonding: allow bond in mode balance-alb to
	work properly in bridge -try4.1

Tue, May 26, 2009 at 06:32:42PM CEST, andy@...yhouse.net wrote:
>On Tue, May 26, 2009 at 05:17:17PM +0200, Jiri Pirko wrote:
>> [PATCH net-next] bonding: allow bond in mode balance-alb to work properly in bridge -try4.1
>> 
>> Hi all.
>> 
>> The problem is described in following bugzilla:
>> https://bugzilla.redhat.com/show_bug.cgi?id=487763
>> 
>> Basically here's what's going on. In every mode, bonding interface uses the same
>> mac address for all enslaved devices (except fail_over_mac). Only balance-alb
>> will simultaneously use multiple MAC addresses across different slaves. When you
>> put this kind of bond device into a bridge it will only add one of mac adresses
>> into a hash list of mac addresses, say X. This mac address is marked as local.
>> But this bonding interface also has mac address Y. Now then packet arrives with
>> destination address Y, this address is not marked as local and the packed looks
>> like it needs to be forwarded. This packet is then lost which is wrong.
>> 
>> Notice that interfaces can be added and removed from bond while it is in bridge.
>> 
>> ***
>> When the multiple addresses for bridge port approach failed to solve this issue
>> due to STP I started to think other way to solve this. I returned to previous
>> solution but tweaked one.
>> 
>> This patch solves the situation in the bonding without touching bridge code.
>> For every incoming frame to bonding the destination address is compared to
>> current address of the slave device from which tha packet came. If these two
>> match destination address is replaced by mac address of the master. This address
>> is known by bridge so it is delivered properly.
>
>Did you test this with a bond with more than 2 ports?  I ask because I
>might also expect a check against all the members of the bond (rather
>than simply the receiving device).

Yes, my testing machine has 3 interfaces for bond. Works fine.
>
>That check would be quite expensive for every frame and I think the
>scenario is quite unlikely based on the frequency of 'learning frames'
>sent by the alb code (so the switch connected to the host should have
>it's forwarding database correct), but it might be something to think
>about in the future.

As you can see, my previous patch did the checking vs all slaves. I tried this
experimentally and searched address from the list and dev->dev_addr differs only
when I unplug cable and mac swap occurs. Then one packet is lost. But there are
many lost packet during the unplug anyway so....

>
>> I experimentally tried that this works as good as searching through the slave
>> list (v4 of this patch).
>> 
>> I was forced to create a new header because I need to use
>> compare_ether_addr_64bits() (defined in linux/etherdevice.h) in
>> linux/netdevice.h. I've hit some cross include issues. I think that it's good
>> to have skb_bond_should_drop() in a separate file anyway.
>> 
>> Jirka
>> 
>> 
>> Signed-off-by: Jiri Pirko <jpirko@...hat.com>
>> 
>
>This certainly won't cure all of the problems that arise with bonding
>and bridging interactions, but it's a step in the right direction.
>
>Acked-by: Andy Gospodarek <andy@...yhouse.net>
>
--
To unsubscribe from this list: send the line "unsubscribe netdev" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html

Powered by blists - more mailing lists