Hi,
I am trying to deploy our SFC OPNFV which is using ODL Boron RC3.5. Very often
the deployment fails because the OVS in the compute or in the controller gets
no openflow tables at all. I have been investigating why and I think I found
the reason and perhaps it is a problem in openflow plugin.
First of all, the br-int brige tries to connect to ODL but it does not as you
can see in this output ("is_connected: true" is missing under br-int bridge):
2d3f2f55-e702-4e68-86ca-96185d1bb5df
Manager "tcp:192.168.0.24:6640"
is_connected: true
Bridge br-int
Controller "tcp:192.168.0.24:6653"
fail_mode: secure
Port "tap3da73bd8-74"
Interface "tap3da73bd8-74"
type: internal
Port "qr-483c4c6e-b3"
Interface "qr-483c4c6e-b3"
type: internal
Port br-int
Interface br-int
type: internal
Port "vxlan-192.168.2.21"
Interface "vxlan-192.168.2.21"
type: vxlan
options: {key=flow, local_ip="192.168.2.22",
remote_ip="192.168.2.21"}
Then, in the logs, I can see that:
1. It tries to change the state of my bridge to BECOMESLAVE , and I can
see things like: RoleChangeTask called on device:openflow:251377037639930
OFPRole:BECOMESLAVE
2. Then there are some Netvirt logs
3. Then, I start seeing worrying messages like: New role BECOMESLAVE was
not propagated to device openflow:251377037639930 during 5 sec and finally Was
not able to set role SLAVE to device on node openflow:251377037639930
When I have it working, I get very quickly after the first step messages like:
"submitRoleChange onSuccess for device:Uri [_value=openflow:88846557454217],
role:BECOMEMASTER", even before the Netvirt logs.
I pasted the logs here: http://pastebin.com/491QFE1w
Have you experienced anything like that? Any guesses what might be happening?
Could that be a bug?
Thanks,
Manuel
_______________________________________________
openflowplugin-dev mailing list
[email protected]
https://lists.opendaylight.org/mailman/listinfo/openflowplugin-dev
