openstack team mailing list archive

[Lars Kellogg-Stedman <lars@xxxxxxxxxxxxxxxx>]

Maybe I sent this out too late at night; I think it slipped below
everyone's radar.  I'm interested in whether or not people think this
behavior is a functional bug, or maybe just a documentation bug:

> I ran into an issue earlier today where I had metadata_host set to the
> *hostname* of our controller.  This got stuffed into an iptables rule
> as...
> 
>   -d os-controller.int.seas.harvard.edu/32
> 
> ...which promptly failed.  Setting this to an ip address fixed this
> particular error, leading me to wonder:
> 
> - Is this expected behavior?
> - Should I always use ip addresses for *_host values?
> - Is this a bug?
> - Should linux_net.py resolve hostnames?

-- 
Lars Kellogg-Stedman <lars@xxxxxxxxxxxxxxxx>       |
Senior Technologist                                | http://ac.seas.harvard.edu/
Academic Computing                                 | http://code.seas.harvard.edu/
Harvard School of Engineering and Applied Sciences |