Re: [pcp] Secure sockets builds have high daemon memory utilisation

[Ken McDonell <kenj@xxxxxxxxxxxxxxxx>]

[trying again ... apologies of others are seeing this message repeated]

Just getting back into the shallow end of the pool after returning from 
darkest Peru ...
On 13/06/14 12:48, Nathan Scott wrote:
>
>
>> ----- Original Message -----
>> ...
> NSS init first -
> ... - avoidance on our part seems the best strategy?
>
> Secondly, the SSL session cache setup - ...
> Making our
> code not call in here at all by default seems like a good option.

The level of bloat here seems to be to be unjustified from any sort of
cost-benefit analysis I can imagine ... PCP needs to be part of solving
performance problems, not part of the problem.

Even if the additional cost is high for the uncommon (today) case of
using secure connections, we should concentrate on getting the common
case back to as close as it was before the secure socket functionality
was added.