Pi-hole drops support for ad blocklists used by browser-based ad-blockers

Discussion in 'other software & services' started by ronjor, Sep 29, 2019.

  1. ronjor

    ronjor Global Moderator

    Joined:
    Jul 21, 2003
    Posts:
    163,072
    Location:
    Texas
  2. 142395

    142395 Guest

    Huh? Why it's mistake when Pi-Hole is a DNS-blocker? Does he confused this with AdGuard?
     
  3. summerheat

    summerheat Registered Member

    Joined:
    May 16, 2015
    Posts:
    2,199
    Well, AdGuardHome is also a DNS blocker which supports the ABP filter syntax which is, indeed, more flexible.

    On the other hand Pi-hole supports regex: If you add, say, doubleclick.net in the Pi-hole Admin Console -> Blacklist and click "Add (wildcard)" this results in the following filter:
    Code:
    (^|\.)doubleclick\.net$
    which blocks all sub-domains as well - similarly to adding
    Code:
    ||doubleclick.net^
    to AdGuardHome.

    But he is right in saying that hosts files are very inefficient if it comes to ad blocking. That's why I'm using Pi-hole with an approach similar to this one.
     
  4. 142395

    142395 Guest

    Yup, I know Pi-Hole but admittedly don't know well about AdGuardHome. But even if it fully supports ABP syntax, that requires MITM capability to be efficient given nowadays most websites are https. And I don't wanna call such a blocker as DNS-blocker, that should rather be called filtering proxy. Or if it only can interpret ABP syntax but apply domain-level blocking (more likely scenario), then there's nothing wrong for Pi-Hole to drop the support.
     
  5. summerheat

    summerheat Registered Member

    Joined:
    May 16, 2015
    Posts:
    2,199
    Well, AdGuardHome only supports a limited set of modifiers. This actually causes problems with filter lists as the $third-party modifier is not supported right now.
     
  6. 142395

    142395 Guest

    That's surely a problem. If domains w/ $third-party are ignored, then filtering capability of the list will be significantly impaired. If blocked, then it will cause many FPs. Anyway if traffic is encrypted it can not distinguish 1st and 3rd party, so I think domain-level blocking is mostly enough for network-level blocker, as long as it supports subdomains as you've explained.
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.