Environment deployment fails if 'swift/api' and 'swift/replication' roles are assigned to the same L3 network (services are listening on the same IP):
2016-02-19 15:53:29 ERROR [838] Error running RPC method granular_deploy: Failed to execute hook 'upload_cirros' command: cd / && ruby /etc/puppet/modules/osnailyfacter/modular/astute/upload_cirros.rb
...
2016-02-19T15:51:19.098799+00:00 err: 2016-02-19 15:51:19.070 4808 ERROR swiftclient [req-0e3f8d69-8c21-4b52-8637-52e982cd14d3 50dc7573961a40b1be8ba3f30cf12f10 04ddb5d036744f29af5908ca40328b62 - - -] Container HEAD failed: http://192.168.0.2:8080/v1/AUTH_04ddb5d036744f29af5908ca40328b62/glance 503 Service Unavailable
2016-02-19 15:51:19.070 4808 ERROR swiftclient Traceback (most recent call last):
2016-02-19 15:51:19.070 4808 ERROR swiftclient File "/usr/lib/python2.7/dist-packages/swiftclient/client.py", line 1390, in _retry
2016-02-19 15:51:19.070 4808 ERROR swiftclient service_token=self.service_token, **kwargs)
2016-02-19 15:51:19.070 4808 ERROR swiftclient File "/usr/lib/python2.7/dist-packages/swiftclient/client.py", line 760, in head_container
2016-02-19 15:51:19.070 4808 ERROR swiftclient http_response_content=body)
2016-02-19 15:51:19.070 4808 ERROR swiftclient ClientException: Container HEAD failed: http://192.168.0.2:8080/v1/AUTH_04ddb5d036744f29af5908ca40328b62/glance 503 Service Unavailable
2016-02-19 15:51:19.070 4808 ERROR swiftclient
...
<129>Feb 19 14:45:31 node-1 haproxy[12097]: Server swift/node-1 is DOWN, reason: Layer4 connection problem, info: "Connection refused", check duration: 0ms. 2 active and 0 backup servers left. 0 sessions active, 0 requeued, 0 remaining in
queue.
<129>Feb 19 14:45:41 node-1 haproxy[12097]: Server swift/node-3 is DOWN, reason: Layer4 timeout, check duration: 10001ms. 1 active and 0 backup servers left. 0 sessions active, 0 requeued, 0 remaining in queue.
<129>Feb 19 14:45:41 node-1 haproxy[12097]: Server swift/node-2 is DOWN, reason: Layer4 timeout, check duration: 10001ms. 0 active and 0 backup servers left. 0 sessions active, 0 requeued, 0 remaining in queue.
...
root@node-1:~# ip netns exec haproxy curl -v 192.168.1.3:49001
* Rebuilt URL to: 192.168.1.3:49001/
* Hostname was NOT found in DNS cache
* Trying 192.168.1.3...
* Connected to 192.168.1.3 (192.168.1.3) port 49001 (#0)
> GET / HTTP/1.1
> User-Agent: curl/7.35.0
> Host: 192.168.1.3:49001
> Accept: */*
>
nc: port number invalid: //192.168.0.2:5000
nc: port number invalid: //192.168.0.2:5000
nc: port number invalid: //192.168.0.2:5000
nc: port number invalid: //192.168.0.2:5000
HTTP/1.1 503 Service Unavailable
There is an error in swiftcheck service config (http://paste.openstack.org/show/487736/), the second arg must have "$ip:$port" format, but currently it has unexpected 'protocol' prefix:
server_args = http://192.168.1.1:8080 http://192.168.0.2:5000 5
Regression was added here: https://github.com/openstack/fuel-library/commit/9d11e9c6567638282813982888c070c39bde6e9e
Steps to reproduce:
1. Create new environment
2. Choose Neutron+VLAN, CinderLVM for volumes. Enable Sahara and Ceilometer.
3. Configure public network (L3 config, do not touch networks to interfaces assignments) and floating ranges through GUI.
4. Add 3 controller+mongo nodes
5. Add 2 compute + cinder nodes
6. Upload network template using CLI:
ENV_ID=1
curl -s https://raw.githubusercontent.com/stackforge/fuel-qa/stable/8.0/fuelweb_test/network_templates/hardware.yaml > /root/network_template_${ENV_ID}.yaml
fuel --env ${ENV_ID} network-template -u
7.Create custom networks using CLI:
GROUP_ID=1
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.0.0/24 --vlan 364 --name ceph
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.1.0/24 --vlan 367 --name database
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.2.0/24 --vlan 366 --name ha
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.3.0/24 --vlan 368 --name messaging
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.4.0/24 --vlan 369 --name mongo
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.5.0/24 --vlan 365 --name openstack
fuel network-group --create --node-group ${GROUP_ID} --cidr 10.44.6.0/24 --vlan 363 --name services
8. Run network verification
9. Deploy environment
Expected result: cloud is deployed and passes OSTF
Actual result: deployment fails, swift backend is permanently marked as down in HAProxy
Diagnostic snapshot: https://drive.google.com/file/d/0BzaZINLQ8-xkR2p1RHU0dklfSGs/view?usp=sharing
Fix proposed to branch: master
Review: https:/