Open Stack

Hi Yipei,

 

I have meant to add this as a config option, but in the interim you can do the following to disable the automatic cleanup by disabling the revert flow in taskflow:

 

octavia/common/base_taskflow.py line 37 add “never_resolve=True,” to the engine load parameters.

 

Michael

 

From: Yipei Niu [mailto:newypei at gmail.com] 
Sent: Monday, June 26, 2017 11:34 PM
To: OpenStack Development Mailing List (not for usage questions) <openstack-dev at lists.openstack.org>
Subject: Re: [openstack-dev] [octavia] fail to plug vip to amphora

 

Hi, Micheal,

 

Thanks a lot for your help, but I still have one question. 

 

In Octavia, once the controller worker fails plugging VIP to the amphora, the amphora is deleted immediately, making it impossible to trace the error. How to prevent Octavia from stopping and deleting the amphora? 

 

Best regards,

Yipei 

 

On Mon, Jun 26, 2017 at 11:21 AM, Yipei Niu <newypei at gmail.com <mailto:newypei at gmail.com> > wrote:

Hi, all,

 

I am trying to create a load balancer in octavia. The amphora can be booted successfully, and can be reached via icmp. However, octavia fails to plug vip to the amphora through the amphora client api and returns 500 status code, causing some errors as follows.

 

                                           |__Flow 'octavia-create-loadbalancer-flow': InternalServerError: Internal Server Error

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker Traceback (most recent call last):

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/usr/local/lib/python2.7/dist-packages/taskflow/engines/action_engine/executor.py", line 53, in _execute_task

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     result = task.execute(**arguments)

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/opt/stack/octavia/octavia/controller/worker/tasks/amphora_driver_tasks.py", line 240, in execute

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     amphorae_network_config)

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/opt/stack/octavia/octavia/controller/worker/tasks/amphora_driver_tasks.py", line 219, in execute

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     amphora, loadbalancer, amphorae_network_config)

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/opt/stack/octavia/octavia/amphorae/drivers/haproxy/rest_api_driver.py", line 137, in post_vip_plug

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     net_info)

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/opt/stack/octavia/octavia/amphorae/drivers/haproxy/rest_api_driver.py", line 378, in plug_vip

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     return exc.check_exception(r)

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker   File "/opt/stack/octavia/octavia/amphorae/drivers/haproxy/exceptions.py", line 32, in check_exception

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker     raise responses[status_code]()

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker InternalServerError: Internal Server Error

2017-06-21 09:49:35.864 25411 ERROR octavia.controller.worker.controller_worker

 

To fix the problem, I log in the amphora and find that there is one http server process is listening on port 9443, so I think the amphora api services is active. But do not know how to further investigate what error happens inside the amphora api service and solve it? Look forward to your valuable comments.

 

Best regards,

Yipei 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20170628/712d1e50/attachment.html>