Ask Your Question

Neutron LBaaS - loadbalancer status and stats return null

asked 2018-01-05 06:34:11 -0500

pathogenferry gravatar image

We have been testing Openstack's LBaaS via the Neutron project. Our version is Mitaka. We can successfully create a load balancer with all the associated listeners, pools, and members, and it works fine. However, we are not able to get the status or stats of the LB, either via the Neutron CLI or the API directly.

Via the CLI:

# neutron lbaas-loadbalancer-stats <id of load balancer>
tuple indices must be integers, not str

# neutron lbaas-loadbalancer-status <id of load balancer>
tuple indices must be integers, not str

and via the API:

# curl -s -H "X-Auth-Token: $TOKEN"  \<id of loadbalancer>/stats

# curl -s -H "X-Auth-Token: $TOKEN"  \<id of loadbalancer>/statuses

I assume that the 'tuple indices' error displayed by the CLI is Python's way of responding to a null value. I have googled extensively for information on this but have found nothing, hence this question.

Many thanks in advance.

edit retag flag offensive close merge delete

1 answer

Sort by ยป oldest newest most voted

answered 2018-01-26 11:05:50 -0500

johnsom gravatar image

I know there were some bugs in neutron-lbaas for stats/status and I'm not sure if those were resolved/impacting in Mitaka or not.

If you are using the reference driver, Octavia, I would check that the health heartbeat messages are making it back to the health manager controller process. Also check you octavia.conf file to make sure the [health_manager] event_streamer_driver is set to queue_event_streamer. This setting tells octavia to sync it's status/stats with neutron-lbaas.


edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower


Asked: 2018-01-05 06:34:11 -0500

Seen: 238 times

Last updated: Jan 26 '18