problem with queue numbers between bridge, bonding and lo devices

Peter Teoh htmldeveloper at gmail.com
Mon Jul 18 19:37:17 EDT 2011


On Sat, Jul 16, 2011 at 9:04 PM, jiangtao.jit <jiangtao.jit at gmail.com>wrote:

> **
> Hi, all
>
> recently, i read some of the Kernel codes
> and i noticed that a bridge and lo device have only one queue when
> allocated
> but a bonding device can have multi queues
> they are all virtual devices
> why they can have diffrent queue numbers
> it confused me a lot
> i would really appreciate if someone could give me some guidelines
> thanks very much
>
>
>

what is the purpose of a "queue"?   reading the documentation:

http://www.kernel.org/doc/Documentation/networking/bonding.txt

<http://www.kernel.org/doc/Documentation/networking/bonding.txt>each queue
correspond to each physical interface "ethX".   and as all these "slave"
interfaces share the same MAC address, packets can be sent out through any
of these interface and received via another - allowing redundancies in terms
of physical links.   multiple interfaces in the same multi-queue all share
the same MAC address.

but the case of bridging is not really for load balancing purposes.
"bridge" literally means bridging two different physical interface, so that
network traffic can hop from one to the other (independent or ignorant of L3
protocol).   each physical interface must have different MAC address, so
that each "queue" is actually ONE interface.

See:

http://www.youtube.com/watch?v=luOFxh1awiA&feature=related

<http://www.youtube.com/watch?v=luOFxh1awiA&feature=related>

>
>
> 2011-07-16
> ------------------------------
> jiangtao.jit
>
> _______________________________________________
> Kernelnewbies mailing list
> Kernelnewbies at kernelnewbies.org
> http://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies
>
>


-- 
Regards,
Peter Teoh
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20110719/d3adc566/attachment.html 


More information about the Kernelnewbies mailing list