maas-devel team mailing list archive

[Dafang Zhang <me2050@xxxxxxxxx>]

Wow this is amazing.  Thanks!  

I just teared down everything and tested your suggested two workarounds.  
The first workaround worked like a charm by simply adding port 80 to the 
maas-server url as the following.  

maas-server: 'http://192.168.77.101:80/MAAS'

Then, "juju deploy" transitioned the service node state from Ready to 
Allocated immediately.


However, the workaround #2 doesn't seem to work, though, as juju is still on version 
0.5+bzr531-0ub.  It's probably already the latest for the Precise line, I guess.

IMHO, MAAS along with juju look very promising.  The frustration for me 
has been the document not being accurate enough.  For a reference, below are the 
instructions took from https://maas.ubuntu.com/docs/juju-quick-start.html.  
The last line should read something like http://hostname.example.com:80/MAAS.  
The added three characters could have saved me lots of hours.  I want to take this 
opportunity to appreciate the responses directly from the dev teams! 

Creating environments.yaml
Create or modify ~/.juju/environments.yaml with the following content:

juju: environments
environments:
  maas:
    type: maas
    maas-server: 'http://localhost:5240'
    maas-oauth: '${maas-api-key}'
    admin-secret: 'nothing'
Substitute the API key from earlier into the ${maas-api-key} slot. You may need to modify the maas-server setting too; if you're running from the maas package it should be something like http://hostname.example.com/MAAS.


----- Original Message -----
From: Julian Edwards <julian.edwards@xxxxxxxxxxxxx>
To: Dafang Zhang <me2050@xxxxxxxxx>
Cc: Maas Devel <maas-devel@xxxxxxxxxxxxxxxxxxx>
Sent: Thursday, November 1, 2012 10:45 AM
Subject: Re: [Maas-devel] Ready State not Changed to Allocated after Juju Deploy

On Thursday 01 Nov 2012 08:30:08 Dafang Zhang wrote:
> In short, yes, the juju packages from quantal seemed fixed it.  
> 
> From what I experienced, I believe the key is to have 12.10 quantal image
> loaded  on the zookeeper node.  This is done by changing the
> /etc/maas/import_pxe_files, executing the maas-import-pxe-files command,
> and using the maas webui to edit the zookeeper node to use 12.10 image.
>  This automatically lifted the juju packages up to version 0.6-1ubuntu1 on
> the zookeeper node.  In the old non-working setup, the zookeeper node was
> loaded with the precise image, and the juju version was 0.5+bzr531-0ub.
> 
> Additional test - I tried both quantal and precise images as the service
> nodes  (for example, mysql is on quantal, tomcat7 on precise), they both
> worked, despite of different version of the juju packages on each node.
> 
> A side note - the maas packages were updated
> from 0.1+bzr1264+ to 0.1+bzr1269+  by pulling them from the
> quantal-proposed repository.
> 
> Best Wishes,

I think I know what's going on here.

Precise is being installed on the bootstrap node. It has the bug where you 
have to specify the port in the URL to the maas server. So when deploying, the 
provisioning agent, which is running the precise version code, will not work 
with the URL that worked fine with the fixed code used to bootstrap in the 
quantal client.

There are two workarounds:
1. add the port to the URL
2. Add:
juju-origin:ppa
to the environments.yaml file so that the bootstrap node uses a different 
(latest) version of the code.

(Or the third workaround is to do what you did)