2 pi holes with gravsync and dual DHCP servers - low % blocked?

Expected Behaviour:

Both piholes should be blocking same % of traffic

Actual Behaviour:

Primary pihole blocks 30% or so of traffic, secondary blocks less than 2%

Debug Token:

(not sure if this is strictly needed, but can provide if needed)

Additional Info

I've got 2 piholes running on my home network. Both run DHCP with different IP address ranges, I've got gravsync keeping their adlists up to date.

But one of them is blocking a much larger % of traffic than the other. To avoid using a DHCP fallback script I've made sure that while bothhave DHCP active, they assign different IP address ranges to their respective devices.

Wondering if there is something off here that I should be worried about?

I think this would almost wholly depends one what clients are connect to which DHCP server and what those specific clients are being used for.

If server A has the Roku, 2 iPads, and Laptop and server B has the fridge and smart thermostat then both both have vastly different traffic passing through them.

I'm not sure you've posted anything more than speculation.

Is there something you want to block cannot?

This is not the expected behavior. With multiple DNS servers available, clients are free to use any of them at any time. Different clients with different browsing history can be using either Pi-hole.

My intent when setting this dual pi-hole up was to make sure that there is a DHCP fallback if one pi fails (and this has happened once so far). This was done instead of creating a DHCP script to switch DHCP on or off on the second pi in the event the first one fails.

I'm not trying to control what connects to what DHCP sever. I'm not too worried about that.

And yes maybe it is speculation. And not really a problem. That's why I'm asking the question.

But to jfb's point, I'd have expected both pi holes to block the same % of traffic. Why would one be less "aggressive" than the other if they are using the same adlists? I understand different devices would use different types of traffic, but beyond that?

It’s not about “aggression.”

If you have two different PiHole serving traffic to different sets of client then it follows that you’re going to see different sets of results.

Again, imagine the scenario.

PiHole A is serving 2 Roku’s, a laptop, and an IPad.

It’s going to have lots of traffic from ads, web browsing, media streaming, iOS games, the App Store, etc.

Let’s say 30% of traffic blocked.

PiHole B is serving a garage door opener, two smart plugs, and smart thermostat.

It’s not going to to have much to block because these are passive devices that may not touch the internet at all.

Let’s say 6% of traffic blocked.

Thus we have different traffic to different PiHoles resulting in different data sets.

To jfb: What you described is exactly what I have going right now. Both for the DNS settings and the DHCP ranges.

I was just concerned that the % blocked was not picking up on the second pi hole.

To jacoby -- I understand what you mean, thank you for clarifying. I noticed that today the two piholes are closer to 30% blocking now. I guess device switch-over to the other made that happen.

I guess I'll mark one of these posts as the solution. As it is working for me, it seems...

Thanks all!

I deleted my reply, since I apparently can't read and you are already doing what I recommended. As you have noted, what you are seeing is normal. Different clients on each Pi-hole result in different block percentages.

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.