[ovs-discuss] switch port numbers seems to change
Rod N. Melton
rnm at pobox.com
Sun Aug 10 16:53:23 UTC 2014
mininet VERSION 2.1.0+
ovs-vsctl (Open vSwitch) 2.1.0
OpenDaylight with OF13
Ubuntu 14.04 in VirtualBox
I have a mininet topology which I created with a python script. I
created
a script which uses ovs-ofctl to add flows to the switches.
The problem which I am having is that it seems that when i exit mininet
and
OpenDaylight and RESTART BOTH PROGRAMS I can see that the mapping
between
interface names (example s3-eth1) and port numbers is CHANGING. This
means that the script I
created for adding flows no longer works. I can see the changes by using
sudo ovs-ofctl show s1
Can someone tell me how to prevent this change when restarting mininet
(or OpenDaylight or Ubuntu)?
Really frustrating.
sudo ovs-ofctl show s3
OFPT_FEATURES_REPLY (xid=0x2): dpid:0000000000000003
n_tables:254, n_buffers:256
capabilities: FLOW_STATS TABLE_STATS PORT_STATS QUEUE_STATS ARP_MATCH_IP
actions: OUTPUT SET_VLAN_VID SET_VLAN_PCP STRIP_VLAN SET_DL_SRC
SET_DL_DST SET_NW_SRC SET_NW_DST SET_NW_TOS SET_TP_SRC SET_TP_DST
ENQUEUE
1(s3-eth2): addr:9e:fc:2c:66:e4:63
config: 0
state: 0
current: 10GB-FD COPPER
speed: 10000 Mbps now, 0 Mbps max
2(s3-eth1): addr:5e:42:96:e9:47:92 << S3-ETH1 WAS PORT 3 LAST TIME I
RAN MININET. WHY DID IT CHANGE??
config: 0
state: 0
current: 10GB-FD COPPER
speed: 10000 Mbps now, 0 Mbps max
3(s3-eth3): addr:1e:b0:98:07:23:c9 << S3-ETH3 WAS PORT 2 LAST TIME I
RAN MININET. why did it change??
config: 0
state: 0
current: 10GB-FD COPPER
speed: 10000 Mbps now, 0 Mbps max
4(s3-eth4): addr:72:86:0e:af:dd:9a
config: 0
state: 0
current: 10GB-FD COPPER
speed: 10000 Mbps now, 0 Mbps max
LOCAL(s3): addr:2a:74:82:09:4a:44
config: 0
state: 0
speed: 0 Mbps now, 0 Mbps max
OFPT_GET_CONFIG_REPLY (xid=0x4): frags=normal miss_send_len=0
Thanks for any help.
Rod
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://openvswitch.org/pipermail/ovs-discuss/attachments/20140810/f27a3ffc/attachment-0002.html>
More information about the discuss
mailing list