Open Stack

On 17 October 2013 10:10, Robert van Leeuwen <
Robert.vanLeeuwen at spilgames.com> wrote:

>
> > The most common cause for something like this is time being out of sync
> or your MQ/DB service > being unreachable.
>
> We changed the name of the dns name for the MQ yesterday but according to
> the logs they all happily connect.
>
>
As I recall the Openstack services can't use a DNS name to connect to
RabbitMQ. This was the case in Essex and I haven't checked with later
versions. Try enabling the management plugin for Rabbit to see whether you
actually are getting any messages moving through the system.

If you are, then you'll have to look elsewhere. The usual process:

 - test basic commands via command-line for each subsystem to see whether
you get errors with any one of them. Start with keystone, then nova, then
the others. If keystone doesn't work, pretty much nothing will work. Try
the commands with '--debug' to get extra clues for where things are going
wrong.
 - if you get errors, are they localised to particular kinds of commands or
does nothing work
 - enable debug logging for a failing service, try a failing command and
see whether you get a clue in the debug log
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack-operators/attachments/20131017/b65590a3/attachment.html>