Open Stack

[redirecting to the mailing list]

On Tue, May 22, 2012 at 5:38 PM, Doug Hellmann
<doug.hellmann at dreamhost.com>wrote:

> Cool. That's not what I was expecting, but I hadn't gotten far enough to
> dig into libvirt yet. We can definitely make cumulative messages work, and
> that does eliminate a lot of my concern about missing a message here or
> there.
>
>
> On Tue, May 22, 2012 at 5:36 PM, James R Penick <penick at yahoo-inc.com>wrote:
>
>>  Libvirt statistics are all cumulative, my collection agent stores them
>> all in a sqlite db, and my monitoring agent compares deltas to determine
>> resource utilization and send alerts.
>>
>> -James
>>
>>
>> -----Original Message-----
>> *From:* Doug Hellmann [doug.hellmann at dreamhost.com]
>> *Received:* Tuesday, 22 May 2012, 2:32pm
>> *To:* James R Penick [penick at yahoo-inc.com]
>> *CC:* Nick Barcet [nick.barcet at canonical.com];
>> openstack at lists.launchpad.net [openstack at lists.launchpad.net]
>> *Subject:* Re: [Openstack] [metering] high-level design proposal
>>
>>
>>
>> On Tue, May 22, 2012 at 4:35 PM, James R Penick <penick at yahoo-inc.com>wrote:
>>
>>> >>-> In addition to a signature, I think we would need a sequence number
>>> >>to be embedded by the agent for each message sent, so that loss of
>>> >>messages, or forgery of messages, can be detected by the collector and
>>> >>further audit process.
>>>
>>> >OK. We have a message id, but I assumed those would be used to eliminate
>>> >duplicates so this sounds like something different or new. It implies
>>> >that the agent knows its own id (not hard) and keeps up >with a sequence
>>> >counter (more difficult, though not impossible).
>>> >Did you have something in mind for how to implement that?
>>>
>>>
>>>  If we're submitting messages every given node with a predictable
>>> frequency, we should be able to determine that a message was lost simply
>>> by noting a gap in the timestamps. Also, if we're sending cumulative
>>> statistics then the loss of a single message (or even a fair number of
>>> them) shouldn't impact our ability to meter too much.
>>>
>>
>>  I don't know if we have cumulative statistics. What does libvirt
>> actually give us?
>>
>>
>>>
>>> -James
>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.openstack.org/pipermail/openstack/attachments/20120522/3477e40f/attachment.html>