Hi,
So, for 5.0, there would be a compile time option to choose c-ares or unbound, or are we considering a clean break?
If we keep both, which one would be considered default? I can imagine unbound when build support is present, otherwise c-ares?
Thanks,
Jaap
> On 21 Aug 2023, at 22:47, Gerald Combs <gerald@xxxxxxxxxxxxx> wrote:
>
> Sounds fine to me. We had overlapping support for c-ares and ADNS for a while, so this isn't new territory. Can you open an issue and set the milestone to "Wireshark 5.x" so this doesn't get lost?
>
> On 8/20/23 12:08 PM, Jaap Keuter wrote:
>> Hi,
>> So we’ve been using the c-ares name resolver for a while now and it’s serving its purpose.
>> However, this is not the only one out there. DNS technologies have evolved somewhat and c-ares does not provide for them.
>> Would it make sense to start looking into using libunbound[1] as a replacement for c-ares to bring these technologies in reach.
>> From a cursory look it seems that the current structure can be retained while shoehorning in unbound.
>> Thoughts? It could be something we could try to achieve for 5.0.
>> [1] https://nlnetlabs.nl/projects/unbound/about/
>> Jaap