Entropy generator in Juju environment?

[Mark Shuttleworth]

On 22/12/11 13:47, Nick Barcet wrote:
> Taking as an hypothesis that the events taking place on the juju
> environment servers depends on enough factor (network speed, machine
> speed and loads, etc), why not continuously use these "time between
> events" as seeds to provide a good source of entropy for all machines in
> the environment? "Time between events" could measure the time it took
> between requests and ack from the juju agent, or other things of this sort.
>
> Any pro's, con's or other ideas on how to solve this in a simpler way?

Never take entropy from a stranger, kids ;-)