Misconfigured HAProxy backend check for aodh
Affects | Status | Importance | Assigned to | Milestone | |
---|---|---|---|---|---|
OpenStack-Ansible |
Fix Released
|
Undecided
|
Daniel Marks |
Bug Description
DISTRIB_ID="OSA"
DISTRIB_
DISTRIB_
DISTRIB_
When deploying the os_aodh role the backends appear as "DOWN" in the haproxy.
backend aodh_api-back
mode http
balance leastconn
stick store-request src
stick-table type ip size 256k expire 30m
option forwardfor
option httplog
option httpchk HEAD /
http-check expect status 401
The HEAD / actually returns "200 OK":
# curl -v http://
* Trying 10.96.224.244...
* Connected to ctr0002_
> GET / HTTP/1.1
> Host: ctr0002_
> User-Agent: curl/7.47.0
> Accept: */*
>
< HTTP/1.1 200 OK
< Date: Wed, 27 Sep 2017 12:31:12 GMT
< Server: Apache
< Content-Length: 449
< Content-Type: application/json; charset=UTF-8
<
* Connection #0 to host ctr0002_
{"versions": {"values": [{"status": "stable", "updated": "2013-02-
The expected status would be returned by a "HEAD /v2" :
# curl --head http://
HTTP/1.1 401 Unauthorized
Date: Wed, 27 Sep 2017 12:33:06 GMT
Server: Apache
x-openstack-
Content-Length: 114
Content-Type: application/json
But as aodh does currently not implement the "/healthcheck" we should stick with "HEAD /" as for most of the other backends.
Fix proposed to branch: master /review. openstack. org/507842
Review: https:/