Hi,
I set up an OpenVSwitch that is connected to a floodlight controller.
Problem is...
(Manually) added flows disappear when connection between a switch and a
controller(floodlight) is disconnected for a short time (about tens
seconds).
Questions are
1) why added-flows in a switch are gone after disconnection from a
controller ?
Isn't it permanently saved when we add flows by "ovs-ofctl add-flow"?
2) What can be problem for disconnecting ?
(If more information should be given, I will post more configuration
data)
3) Are there any ways to re-add flows automatically when connection is
re-established to a controller ?
----------------------------------------
Below, I show the commands that I used to add flows.
Also, following syslog shows that a connection between a switch and a
controller was disconnected and then is connected some time later (tens
seconds). Then, added flows are gone.
>>> Add flows
root@server:~# ovs-vsctl show
19969e3c-269f-4186-ad4c-ea0212c92b92
Bridge "br0"
Controller "tcp:192.168.1.1:6633"
is_connected: true
fail_mode: secure
Port "eth2"
Interface "eth2"
Port "br0"
Interface "br0"
type: internal
Port "eth1"
Interface "eth1"
root@server:~# ovs-ofctl add-flow br0 in_port=2,actions=output:LOCAL,1
root@server:~# ovs-ofctl add-flow br0 in_port=1,actions=output:LOCAL,2
root@server:~# ovs-ofctl dump-flows br0
NXST_FLOW reply (xid=0x4):
cookie=0x0, duration=6.440s, table=0, n_packets=0, n_bytes=0, idle_age=6,
in_port=1 actions=LOCAL,output:2
cookie=0x0, duration=12.855s, table=0, n_packets=2, n_bytes=130,
idle_age=9, in_port=2 actions=LOCAL,output:1
>>> Logs
- /var/log/syslog
Jan 16 09:25:31 server kernel: [77134.967321] device eth0 entered
promiscuous mode
Jan 16 09:27:10 server ovs-vswitchd: ovs|01256|rconn|ERR|br0<->tcp:
192.168.1.1:6633: no response to inactivity probe after 5 seconds,
disconnecting // <--- disconnect
Jan 16 09:27:11 server ovs-vswitchd: ovs|01257|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connecting...
Jan 16 09:27:12 server ovs-vswitchd: ovs|01258|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connection timed out
Jan 16 09:27:12 server ovs-vswitchd: ovs|01259|rconn|INFO|br0<->tcp:
192.168.1.1:6633: waiting 2 seconds before reconnect
Jan 16 09:27:14 server ovs-vswitchd: ovs|01260|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connecting...
Jan 16 09:27:16 server ovs-vswitchd: ovs|01261|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connection timed out
Jan 16 09:27:16 server ovs-vswitchd: ovs|01262|rconn|INFO|br0<->tcp:
192.168.1.1:6633: waiting 4 seconds before reconnect
Jan 16 09:27:20 server ovs-vswitchd: ovs|01263|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connecting...
Jan 16 09:27:24 server ovs-vswitchd: ovs|01264|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connection timed out
Jan 16 09:27:24 server ovs-vswitchd: ovs|01265|rconn|INFO|br0<->tcp:
192.168.1.1:6633: continuing to retry connections in the background but
suppressing further logging
Jan 16 09:27:39 server ovs-vswitchd: ovs|01266|rconn|INFO|br0<->tcp:
192.168.1.1:6633: connected // <--- connected again
Jan 16 09:27:49 server ovs-vswitchd: ovs|01267|ofproto|INFO|br0: 1
flow_mods 10 s ago (1 deletes)
- floodlight Web UI
*** no flows exist
- check flows that I added
*** no flows exist
root@server:~# ovs-ofctl dump-flows br0
NXST_FLOW reply (xid=0x4):
Danny
_______________________________________________
openflow-discuss mailing list
openflow-discuss@lists.stanford.edu
https://mailman.stanford.edu/mailman/listinfo/openflow-discuss
