Senlin Installation problems

asked 2020-05-10 05:11:52 -0600

AB239 gravatar image

updated 2020-05-10 07:08:08 -0600

Hello,

I have a existing functional setup with 1 controller and 2 compute hosts on Openstack STEIN. I want to test scaling feature of openstack (which I had tested on Newton with HEAT templates) and found that Openstack has moved to SENLIN for scaling feature-set.

There is no installation steps for Ubuntu based OS whereas steps for CENTOS exist. So, I tried to follow the same procedure with obvious changes in commands on ubuntu 18.04 and managed to install and bring senlin-api and senlin-engine up and running.

senlin   12210     1  1 15:32 ?        00:00:04 /usr/bin/python3 /usr/bin/senlin-api --config-file=/etc/senlin/senlin.conf --log-file=/var/log/senlin/senlin-api.log
senlin   12327     1  0 15:32 ?        00:00:03 /usr/bin/python3 /usr/bin/senlin-engine --config-file=/etc/senlin/senlin.conf --log-file=/var/log/senlin/senlin-engine.log

But, when I run openstack cluster build info, I get the below pasted error:

openstack cluster build info 
Failed to contact the endpoint at http://controller:8777/ for discovery. Fallback to using that endpoint as the base url.
Failed to contact the endpoint at http://controller:8777/ for discovery. Fallback to using that endpoint as the base url.
The clustering service for : exists but does not have any supported versions.

Please let me know how I can resolve this problem. There is very limited information available on internet about this error.

edit retag flag offensive close merge delete

Comments

I know nothing about Senlin, but "Failed to contact the endpoint" looks like the Senlin API service doesn't reply. Is it running? Is the port protected by a firewall? Is something wrong with the web server configuration?

I would also check whether any traffic to port 8777 arrives at the controller.

Bernd Bausch gravatar imageBernd Bausch ( 2020-05-10 06:23:55 -0600 )edit

@Bernd: I checked that earlier &8777 is not listening. I don't see any errors in syslog/senlin-engine.log. I also went ahead & re-created endpoints with port 8778 (which is default for senlin-api) but the error changes to TypeError: request() got an unexpected keyword argument 'global_request_id'

AB239 gravatar imageAB239 ( 2020-05-10 06:46:59 -0600 )edit

Are there any errors in syslog/senlin-api.log when you see global_request_id error?

dtruong gravatar imagedtruong ( 2020-05-11 16:34:22 -0600 )edit

@dtruong: 2020-05-11 19:15:46.544 25262 INFO eventlet.wsgi.server [req-27536525-4064-467f-97b8-4bb45ef6b158 deea288913794e90aaa1c5452afec142 707ae77cbc7749ce934578a509e80165 - - -] 10.0.1.28 - - [11/May/2020 19:15:46] "GET /v1/build-info HTTP/1.1" 500 509 61.886461

AB239 gravatar imageAB239 ( 2020-05-12 06:08:04 -0600 )edit

Is the senlin-engine running? Also, you can try setting log level debug in senlin.conf to see more information.

dtruong gravatar imagedtruong ( 2020-05-12 11:58:55 -0600 )edit

1 answer

Sort by ยป oldest newest most voted
0

answered 2020-05-12 23:12:01 -0600

AB239 gravatar image

Issue has been fixed. There were lot of issues that we have unearthed. We remove apt version of senlin-api and engine and installed using source code.

  1. Created virtuanenv with python3
  2. Upgraded requests package in virtualenv
  3. Started senlin-conductor and health manager from virtualenv
  4. Run all cluster related commands from virtualenv

This should fix some initial issues as they solved for me.

edit flag offensive delete link more

Get to know Ask OpenStack

Resources for moderators

Question Tools

1 follower

Stats

Asked: 2020-05-10 05:11:52 -0600

Seen: 168 times

Last updated: May 12