<br><br><div class="gmail_quote">On Wed, Nov 28, 2012 at 6:13 AM, Julien Danjou <span dir="ltr"><<a href="mailto:julien@danjou.info" target="_blank">julien@danjou.info</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<div class="im">On Tue, Nov 27 2012, Doug Hellmann wrote:<br>
<br>
>> I may have a better idea.<br>
>> The problem with doing that in the pollster, is that:<br>
>> 1. it's bad bad bad bad<br>
>> 2. it's inefficient since you may lose values on restart<br>
>> you won't know what the value was before<br>
>><br>
><br>
> We will if we persist them on the compute node.<br>
<br>
</div>See point 1. :-)<br>
<div class="im"><br>
> That doesn't work if different collector instances receive events from the<br>
> same meter on the same resource in parallel (which could happen easily if<br>
> the queue gets backed up).<br>
<br>
</div>It does if you don't try to cache locally but request the last value<br>
>From the storage backend, right?<br></blockquote><div><br></div><div>There's a race condition, though. If you and I are processing events at the same time, we get the same "last value" from the storage but then try to update it to our current value.</div>
<div><br></div><div>Doug</div><div> </div><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
<span class="HOEnZb"><font color="#888888"><br>
--<br>
Julien Danjou<br>
// Free Software hacker & freelance<br>
// <a href="http://julien.danjou.info" target="_blank">http://julien.danjou.info</a><br>
</font></span><br>_______________________________________________<br>
OpenStack-dev mailing list<br>
<a href="mailto:OpenStack-dev@lists.openstack.org">OpenStack-dev@lists.openstack.org</a><br>
<a href="http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev" target="_blank">http://lists.openstack.org/cgi-bin/mailman/listinfo/openstack-dev</a><br>
<br></blockquote></div><br>