Comment 2 for bug 1055737

Reviewed: https://review.openstack.org/13594
Committed: http://github.com/openstack/nova/commit/8887f10c66bca248f289db8f834ae8f36f9a03a1
Submitter: Jenkins
Branch: master

commit 8887f10c66bca248f289db8f834ae8f36f9a03a1
Author: Monsyne Dragon <email address hidden>
Date: Mon Sep 24 19:38:09 2012 +0000

    Collect more accurate bandwidth data for XenServer

    This changes the method used to poll xenserver for bandwidth data.
    The reccomended way of collecting such data from xenserver (namely the
    RRD files provided by the hosts) do not seem to be reliable, they
    will sometimes be correct, often will be signifigantly under (> 10%),
    and occasionally will show artifacts, such as phantom 4gb bandwidth
    'spikes'.

    This patch changes that to use the much simpler method of simply polling the
    byte counters on the VIF network devices on the host. (We have old non-nova
    code that does that on xenserver, and that method is known to work).

    This should also make it much easier for other hypervisors other than
    xenserver to implement bandwidth polling, as polling the counters is a rather
    more universal method.

    Fixes bug 1055737

    Change-Id: I6a280d8bbfcc74914f888b11bc09349a270a5f58