Open Stack

> What sort of resolutions are we talking here? Are we talking instrumentation
> levels (like 1/min?) or usage event levels (perhaps as fast as every 10 minutes)?
> 
> Regardless, could you provide some info on the services that are suffering so
> badly that eventlet lags the periodic_task for so long? Sounds like a nova bug
> we should fix? Actually, if we're monitoring green thread counts via the periodic
> task we should be able to warn upstream monitoring before it becomes an
> issue.
> 
> (mixing real threads with eventlet will be a sure-fire p.i.t.a)

I remember it's about several seconds level, but I didn't find it in this big thread. And I think that make sense considering auto-scaling situation.

--jyh