Legit domains are being blocked in Pihole v5

Help Community Help
1

I have just installed a new instance of Pi-hole v5 and I immediately noticed that I am getting a large number of blocked domains that are legit. Specifically, things like google.com, github.com, www.apple.com, ..., etc. I am having to whitelist each manually. My Pi-hole v4 instance does not do this.

I read the documentation and nothing stood out. Is there a way to reduce the false-positives? This behavior renders Pi-hole useless, as I am having to constantly whitelist domains.

Any pointers would be greatly appreciated.

2

Please post the token generated by

pihole -d

or do it through the Web interface:

Tools > Generate Debug Log

3

https://tricorder.pi-hole.net/ocamk3o6m8

Thank you

4

Second Pi-hole instance:

https://tricorder.pi-hole.net/7a4svzsw47

Thank you

5

It looks like I solved my own issue, I removed all the Adlist lists that I imported from my v4 instance and kept only the 4 Adlists that came with Pi-hole 5. Now the raw domains, like github.com are not being blocked. This issue does not happen on v4 instance. The question is:

Are the Adlists such as https://easylist.to/easylist/easylist.tx not compatible with v5?

I may need to import them one-at-a-time and see which one is causing the problem.

Thank you

8

They are not compatible in either V4 or V5, but importing them is handled differently in V5. They are not in HOSTS format. If you look at the output from a gravity rebuild (pihole -g), you will see the following:

 [i] Target: https://easylist.to/easylist/easylist.txt
  [✓] Status: Retrieval successful
  [i] Received 33321 domains, 22804 domains invalid!
      Sample of invalid domains:
      - 2.0]
      - (easylist.subscription@gmail.com).
      - &ad.vid=$~xmlhttprequest
      - &advertiserid=$domain=~bee.gl|~cpashka.ru
      - -ad.jpg?

Even if some of the domains are parsed, they are likely to be imported not as intended to be used in that list. As shown below

pihole -q google.com

 Match found in https://easylist.to/easylist/easylist.txt:
   transparencyreport.google.com 
   maps.google.com 
   google.com.au 
   google.com 
   google.com 
   google.com 
   google.com 
   google.com 
   mail.google.com 
   maps.google.com 
   maps.google.com 
   google.com 
   mail.google.com 
   google.com 
   accounts.google.com 
   www.google.com

There is no reliable way to parse these lists, and using them causes problems as you have seen.

9

Ah ... Ok, that makes sense and explains the oddity.

Thank you for the helpful information!