Ask Your Question
0

Loadbalancer agent shown as not alive

asked 2014-06-23 00:11:50 -0500

vyom gravatar image

updated 2014-07-01 01:42:43 -0500

Hello,

After the reboot of my host (--allinone), I see that Loadbalancer agent state is shown as xxx (not alive), even though neutron-lbaas-agent is running. I see the following in the log though:

Timeout: Timeout while waiting on RPC response - topic: "n-lbaas-plugin", RPC method: "get_ready_devices" info: "<unknown>"

rabbitmq is also up and running, and I have the following in my neutron.conf

report_interval = 30
agent_down_time = 75

what else should I check?

--EDIT-2014/7/1--

Even with state being shown xxx, the functionality is not affected.

edit retag flag offensive close merge delete
add a comment

1 answer

Sort by ยป oldest newest most voted
0

answered 2014-06-23 01:33:20 -0500

KunalTPatil gravatar image

LBAAS agent is not enabled when you use RDO Packstack to install OpenStack Icehouse.

I used the steps from http://docs.openstack.org/admin-guide-cloud/content/install_neutron-lbaas-agent.html (http://docs.openstack.org/admin-guide...) to enable it on the network node. Replace apt-get with yum (yum install neutron-lbaas-agent). Also make sure haproxy is installed (yum install haproxy).

Run the following to enable the service (after httpd and neutron services are restarted):

service neutron-lbaas-agent start

chkconfig neutron-lbaas-agent on

edit flag offensive delete link more

Comments

@Kunal, I had HaProxy installed and running, and then I replaced it with my own driver. Even with my driver, I had the agent status as up. After a host reboot, I am seeing this issue.

vyom gravatar imagevyom ( 2014-06-23 04:16:38 -0500 )edit

Similar issue faced earlier: https://ask.openstack.org/en/question/13226/lbaas-unable-to-retrieve-ready-devices/ (https://ask.openstack.org/en/question...) . You may try a few things from here. What happens when you restart the neutron-lbaas-agent service? Also, it would help to get extra info in logs by setting debug true in the conf files.

KunalTPatil gravatar imageKunalTPatil ( 2014-06-23 04:40:19 -0500 )edit

Check for this bug. I think it might help: https://bugzilla.redhat.com/show_bug.cgi?format=multiple&id=1022725 (https://bugzilla.redhat.com/show_bug....) [Assuming there is some error in the /var/log/neutron/server.log file as well]

KunalTPatil gravatar imageKunalTPatil ( 2014-06-23 22:47:15 -0500 )edit

Summarised steps. I faced similar problems in the morning. There are some bugs you need to work around. http://itinsteps.blogspot.in/2014/06/steps-to-configure-neutron-lbaas-agent.html (http://itinsteps.blogspot.in/2014/06/...)

KunalTPatil gravatar imageKunalTPatil ( 2014-06-24 00:26:25 -0500 )edit

@Kunal, as mentioned earlier, I am not using HAProxy; I have my own LBaaS driver. Also, I dont have service_provider config in neutron-dist.conf

vyom gravatar imagevyom ( 2014-06-24 12:01:49 -0500 )edit
add a comment

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

subscribe to rss feed

Stats

Asked: 2014-06-23 00:11:50 -0500

Seen: 624 times

Last updated: Jul 01 '14

Related questions

cinder "Failed to create iscsi target for volume"

Icehouse RDO install error: Error: /Stage[main]/Keystone::Db::Sync/Exec[keystone-manage db_sync]: Failed to call refresh: keystone-manage db_sync returned 1 instead of one of [0]

Why should packstack be run as regular user and not as root?

Upgrading RDO To Icehouse on Fedora 20

magnum create cluster volumeType error

internal server 500 dashboard icehouse

Paid Support

icehouse on fedora21

packstack installation fails with mysql access denied error

trusty apt-get update update fails on http://ubuntu-cloud.archive.canonical.com [closed]