[gsoc-dev] (no subject)

Lance Albertson lance at osuosl.org
Tue May 7 19:18:50 UTC 2013


On Tue, May 7, 2013 at 11:45 AM, pranjal mittal <mittal.pranjal at gmail.com>wrote:

> Ok, got the secondary_node information. As you said, the raw config file
> does not seem to organize things well. Maybe that could be improved with
> Ganeti.
>

I don't think this really matters since its more about machine code being
able to read it. Its more important to make sure the RAPI is a good
standard.


> Just wrote up a small script to extract out  some of the useful part from
> this config file:
>
> https://github.com/pramttl/ganeti-cluster-visualisation/blob/master/extract_info.py
>
> *instance_name*, *primary_node*, *secondary_node*
> (This is the most basic information required initially for the graphical
> models.)
>

Excellent! I ran it and it seems to work fairly well. I realized that we
don't really use node groups so unfortunately so I can't give you a good
example for that. Maybe you can try and mimic that in a vagrant ganeti
cluster somehow with just three nodes. Or you could email the Ganeti
community and ask them for help but I'll leave that up to you. For a
reference, I used the *sanitize-config* script in the *tools* directory.

-- 
Lance Albertson
Director
Oregon State University | Open Source Lab
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osuosl.org/pipermail/gsoc-dev/attachments/20130507/bd3732d2/attachment.html>


More information about the gsoc-dev mailing list