[ovs-discuss] ARP Behavior in XenServer Host

David Erickson derickso at stanford.edu
Fri Sep 9 01:11:39 UTC 2011

Hi All-
I'm seeing some confusing behavior happening relating to ARPs and OVS 
1.2.1 (also on 1.1.1 and probably 1.0.1). This is a XS machine with 1 
ethernet port, so in-band connection from OVS to the controller. 
Specifically there are two cases where I would expect packets to be sent 
up to the connected controller, but they aren't:

1- VLAN Tagged ARP replies coming in eth0, attached as tagged_arp.pcap

2- Gratuitous ARPs sent from the XS host, eg: /sbin/arping -U -c 1 -I 
xenbr0 -s, attached as gratuitous_arp.pcap

As a sanity check the flows installed into the xenbr0 datapath are the 
following, which should not be matching either of these types:

[root at host0005 ~]# ovs-ofctl dump-flows xenbr0
NXST_FLOW reply (xid=0x4):
  cookie=0x546, duration=528.285s, table=0, n_packets=384, 
  cookie=0x547, duration=528.25s, table=0, n_packets=520, n_bytes=46728, 

Also is there a document that specifies OVSs behavior in in-band 
environments such as XS with one ethernet port?  I would expect after 
the controller is connected that all packets coming in/out other than 
the controller connection (and possibly ARPs towards the controller 
ip/gateway?) should flow through the flowtable/controller, but maybe my 
expectation is incorrect.

-------------- next part --------------
A non-text attachment was scrubbed...
Name: tagged_arp.pcap
Type: application/cap
Size: 100 bytes
Desc: not available
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20110908/859367be/attachment.bin>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: gratuitous_arp.pcap
Type: application/cap
Size: 82 bytes
Desc: not available
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20110908/859367be/attachment-0001.bin>

More information about the discuss mailing list