libcloud-notifications mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Anthony Shaw (JIRA)" <>
Subject [jira] [Commented] (LIBCLOUD-912) [python] Requests tries to connect to
Date Fri, 21 Apr 2017 06:18:04 GMT


Anthony Shaw commented on LIBCLOUD-912:

thanks for raising this. We need to get it fixed.

Can you give me a code snippet about how you instantiate the driver and what arguments you
use so I can reproduce it?

When you say Rackspace/openstack are you using the rackspace driver or the openstack one directly?

The behaviour you're seeing is because the OpenStack driver creates 2 connections, one to
the API and one to the authentication endpoint, the second is created after instantiation
of the first. If you see, thats the default host address (although it should be
none or something more obvious that it's broken).
I assume what's happening is the host isn't being passed to the second connection class. 

> [python] Requests tries to connect to
> ---------------------------------------------------
>                 Key: LIBCLOUD-912
>                 URL:
>             Project: Libcloud
>          Issue Type: Bug
>         Environment: python 2.7
> apache-libcloud==2.0.0rc2
>            Reporter: Alexis Lesieur
>              Labels: newbie, python
> Hi!
> Sorry if this story is not created properly.
> I'm not that familiar with how you guys manage your Jira. Feel free to give me directions
or just modify the ticket :-)
> So we have been trying to upgrade the version of our libcloud python package, but it
breaks for our rackspace/openstack environment.
> I have been digging for a while and I couldn't find where the difference was between
1.x.x and 2.0.0rc2.
> Observed behavior:
> 1.x.x:
> 3 connections are created:
> - to
> - to host:3000 (ex_force_auth_url)
> - to host:8774/v2/... (don't know where it comes from)
> 2.0.0rc2:
> 2 connections
> - to
> - to host:3000 (ex_force_auth_url)
> and the last connection is done using again, using the same connector created
in the first one
> I don't understand the change in behavior between the two versions.
> I tried use the ex_force_base_url flag, and while it does look like the path of that
url is respected, the host is always replaced by
> Any idea of what is happening?
> Thanks!

This message was sent by Atlassian JIRA

View raw message