On 1/29/25 4:25 PM, Theo de Raadt wrote:
>> MAXNS is the numebr of NS actually used from the list.
>>
>> I don't think upping MAXNS to 5 is useful, as the timeouts will
>> accumulate to very high values.
>>
>> We only implement try in-order, but some implementations allow
>> randomization of NS used. In that case the # of listed and the number
>> used can differ in a meaningful way.
> 
> Right.
> 
> I guess the question is why does the OP want a different number.
> 
> The value is not publically exposed, because noone should code to it.
> It's a detail that applications should not interact with.  It's not
> part of the interface, and the behaviour of that number is also not in
> play.

Actually, I didn't specifically want to increase or decrease any of 
those values. I just wanted to know if it was normal that there were 
different values between internal libc ASR_MAXNS and resolv.h MAXNS.

• smime.p7s (5K)