Share wildcard blacklists

Hello,

This is my first week of using pi-hole, so bear with me if I am asking for something that can already be accomplished :slight_smile:

Mind you, I love the public blacklists tat can be added but they are all in /etc/hosts format. A quick glance of the (very nice) dashboard shows that this is not sufficient.
For example the TLD of some bitcoin mining programs are blocked, while “randomly named” sub domains are still left unblocked (ws37a.work.miner... etc).

So now I keep an eye out to these kind of entries in the querie log (where the TLD is blocked, but suspicious domains still resolve through a sub domain - of which an infinite amount can exist per given domain).

What I would like is a way to share one’s wildcard blacklist and import other people’s wildcard blacklist. Exactly like it now works with /etc/hosts formatted lists.

Again, I am new to this game, and it might involve a lot of politics as other adblockers probably need to jump on the train as welll. Just putting this wish up for discussion.

Edit: if I owned a miner or adfarm (which I do not) I would use a wildcard DNS record for my clients to bypass this type of blocking. Community wildcard blocking would help IMHO.

1 Like

This is possible, but so far we have not implemented it as a feature (might be a way to get involved though, if you know how to code :wink:)

Hmm, I just might do that :slight_smile:.

But first, let’s think about the “distributable wildcards lists” themselves.
How will they be generated? How to categorize them (is that necessary)?
@WaLLy3K, you seem to collect these lists, how did they came to be? Any tips/requirements on how wildcard list should be organized?

For the content of the lists I obviously can share my own, but that will mostly block stuff that is pulled in by whatever freemium game the missus plays this week :grin:. Automating the creation does not seem possible (maybe for the mining category we can just use all TLDs found in those lists, but for other lists the risk of blocking Apple /Google/Microsoft etc entirely seems too high if done automatically).

Once you have the domains, it's the same process as the current lists. Just the domains would be added to the wildcard list instead of the blacklist.

Maybe I am just overthinking it, but if this feature is to be used we need meaningful wildcard blacklists. These have to be provided by users of pi-hole (unlike the existing host blacklists, which are the results of years of gathering from different sources, compiled by different methods and by a wild variety of audiences - I assume). As wildcard blacklists need to be build up from the ground, I was looking for requirements on stuff like the GUI to enable pi-hole users to compile such lists with ease (and since user prefer GUIs which wont work for large lists and since I think lists are better if they are categorized for example, it really needs to be thought through before implementing).

The other way would be to just use either existing host lists directly as wildcard list (with some filtering to remove duplicates) or use the existing hosts list and determine which A-records have a wildcard and put that on the wildcard blacklist. Both seem possible, not sure on dnsmasq performance with huge wildcard lists yet.

Gonna see if I can find the dnsmasq limits with wildcards next.

Closing as duplicate in favor of

as both request the ability to import wildcard/regex lists. Internally pihole treats wildcard domains as regex domains.
Votes are released. Please vote for the other FR if you still support the request.