[ovs-discuss] openvswitch noraml behaviour

Vahric Muhtaryan vahric at doruk.net.tr
Sun Jan 3 12:40:31 UTC 2016


Hello , 

When you create a bridge like below then you will have standard switch , it
will behave like L2 switch , it will learn MAC and make forwarding for
virtual servers communications .

sudo ovs-vsctl add-br learning-switch

Default OVS switch ports are Trunk and you can also specify access ports for
virtual servers like below and also see the MAC Table like traditional
switch

   <interface type='bridge'>
 <mac address='52:54:00:a0:c1:81'/>
 <source bridge='bridge1'/>
 <vlan>
         <tag id='497'/>
       </vlan>
 <virtualport type='openvswitch'/>
 <parameters interfaceid='4b25e7e1-7db2-4133-a8b8-0c050f4983d2'/>
</virtualport> 

> noroot at kvm-ovs-server1:~$ sudo ovs-appctl fdb/show learning-switch
> 
> port  VLAN  MAC                Age
> 
> 4   497  52:54:00:6c:74:56    0
> 
> 2   497  52:54:00:bd:d1:a7    0

Also you can enable/disable STP too !

Regards
VM

From:  discuss <discuss-bounces at openvswitch.org> on behalf of ?Mohammed
Abuzaid? ? <mohmd_abz at yahoo.com>
Reply-To:  ?Mohammed Abuzaid? ? <mohmd_abz at yahoo.com>
Date:  Sunday 3 January 2016 at 13:59
To:  "discuss at openvswitch.org" <discuss at openvswitch.org>
Subject:  [ovs-discuss] openvswitch noraml behaviour

hi , everyone

i'm doing a performance comparison between SDN and a traditional network
using mininet, but unfortunately mininet doesn't emulate traditional
networks directly.

some solutions suggests using openvswitch in the normal switch behavior by
installing flow entry with "Normal Forwarding" , is that correct ? is there
another way to let the switch completely behaves as A tradition switch ?
e.g ,,  Vlans ,Spanning Tree ?
_______________________________________________ discuss mailing list
discuss at openvswitch.org http://openvswitch.org/mailman/listinfo/discuss

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20160103/175fb77a/attachment-0002.html>


More information about the discuss mailing list