$ neutron port-show 960ea43e-5644-47f8-ab30-a6cdd6dde664
+-----------------------+---------------------------------------------------------------------------------------+
| Field | Value
|
+-----------------------+---------------------------------------------------------------------------------------+
| admin_state_up | True
|
| allowed_address_pairs |
|
| binding:host_id | node-A
|
| binding:profile | {}
|
| binding:vif_details | {"port_filter": true, "ovs_hybrid_plug": false}
|
| binding:vif_type | ovs
|
| binding:vnic_type | normal
|
| created_at | 2016-05-27T01:24:32
|
| description |
|
| device_id | a54ffb2d-bbfe-4c4b-85be-5c09140273c2
|
| device_owner | network:router_ha_interface
|
| dns_name |
|
| extra_dhcp_opts |
|
| fixed_ips | {"subnet_id": "097af809-cd4b-4801-adf4-26ce04584791",
"ip_address": "169.254.192.20"} |
| id | 960ea43e-5644-47f8-ab30-a6cdd6dde664
|
| mac_address | fa:16:3e:c0:6f:24
|
| name | HA port tenant a17b3b9772b74811a3d62ebefd65d3b0
|
| network_id | dc3c02c5-c1c5-4950-83db-939740e73916
|
| port_security_enabled | False
|
| security_groups |
|
| status | ACTIVE
|
| tenant_id |
|
| updated_at | 2016-05-27T02:28:02
|
+-----------------------+---------------------------------------------------------------------------------------+
$ ip netns exec qrouter-a54ffb2d-bbfe-4c4b-85be-5c09140273c2 ping 169.254.192.21
PING 169.254.192.21 (169.254.192.21) 56(84) bytes of data.
>From 169.254.192.20 icmp_seq=1 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=2 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=3 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=4 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=5 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=6 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=7 Destination Host Unreachable
>From 169.254.192.20 icmp_seq=8 Destination Host Unreachable
@Assaf Muller, it seems strange, I do another test in other node, stop l3
agent, create router(both ha and ha+distributed), then start l3 agent, it seems
OK, there only one active ha router, and vip address only in one namespace.
Thanks for your kindly reply, I will do more tests.
** Changed in: neutron
Status: Incomplete => Invalid
--
You received this bug notification because you are a member of Yahoo!
Engineering Team, which is subscribed to neutron.
https://bugs.launchpad.net/bugs/1586058
Title:
Online an new neutron l3 agent, the router ha_state is active
Status in neutron:
Invalid
Bug description:
When we manually online an new l3 agent, after sync router info from
Neutron server, neutron-keepalived-state-change will always write
master in it's state file and notify other agents, even though there
still have an active ha router in other node.
====== Step to reproduce =======
1. stop neutron-l3-agent on node-A
2. create ha and distribute router test_router
3. we can see l3 agents hosting router info:
$ neutron l3-agent-list-hosting-router test_router
+--------------------------------------+------------+----------------+-------+----------+
| id | host | admin_state_up | alive
| ha_state |
+--------------------------------------+------------+----------------+-------+----------+
| 678cc6a6-3b28-462d-9374-4a7fb42193dc | compute1 | True | :-)
| standby |
| de274597-e35b-4ae5-baba-01c26bbed5b9 | compute2 | True | :-)
| active |
| 1c301ec4-9320-44e0-a4e0-a0441fee849a | network | True | :-)
| standby |
| 4ae490f7-8791-4460-8545-c172af58de8f | network2 | True | :-)
| standby |
+--------------------------------------+------------+----------------+-------+----------+
4. start neutron-l3-agent on node-A, after sync, we can see there are
two active ha routers in compute2 and node-A.
To manage notifications about this bug go to:
https://bugs.launchpad.net/neutron/+bug/1586058/+subscriptions
--
Mailing list: https://launchpad.net/~yahoo-eng-team
Post to : yahoo-eng-team@lists.launchpad.net
Unsubscribe : https://launchpad.net/~yahoo-eng-team
More help : https://help.launchpad.net/ListHelp
