Discuss mailing list

[Mike <mike AT pikeaero.com>]

Nothing like waiting for a kettle to boil!... :)

Some interesting observations though as I'm waiting for CPU usage to
explode....I have the CPU graph running, as well as the OpenNIC applet
Journal open.

In the initial half an hour or so, I observed fairly frequent changing
of the active resolvers.

However as time goes on and the average latencies get better
established, the changing becomes less and less frequent, exactly as I
intended. It's just interesting to want it actually working as intended.
Each candidate in the resolver pool has a sample queue 10 elements deep
from which to draw an average.

Sample intervals are random per resolver in the pool, and the interval
gets re-randomized at the expiration of the previous interval. In other
words, each resolver in the queue has it's own independent sampling context.

Anyway just thought I'd share those observations while I wait for
something to explode here.

--Mike

On 02/28/2012 11:24 PM, Mike wrote:
> Ahhh..maybe not so fast with the beers, my friend!
> 
> I have had 0.2.4 running for about 4 hours on Windows 7 and suddenly CPU
> usage shot up to %100 and it is none other than OpenNIC.exe and
> OpenNICServer.exe consuming about %50 each.
> 
> So I've got them both running in the debugger now to wait for it to
> happen again so I can see what's letting loose there.
> 
> So something to keep an eye out for if you're running 0.2.4.
> 
> --Mike
> 
> On 02/28/2012 08:43 PM, Hospedaje Web y Servidores Dedicados wrote:
>> So now are two beers...
>>
>> :p
>>
>>
>