Just a little public service announcement here. When you are setting up your fancy DNS services on TrueNAS (Pi-hole, I’m looking at you), please do not point your TrueNAS host at it that DNS. This creates a nasty loop, where your TrueNAS services won’t start because DNS isn’t responding, and your services that host your DNS now will never start as a result.
The same goes for hosting AD directly on your TrueNAS device. You can host services, but you want to ensure you aren’t making the TrueNAS host depend on those services being “up” for it to complete its own boot-up routines
The only way to recover from this is to override the boot (or boot environment) via the “console menu”? (Lest you do a brand new installation and reconfigure your TrueNAS system?)
I’ve embarrassingly “soft bricked” my tomato-flashed router one time by seriously messing up WAN and VPN settings, essentially causing a loop.
Luckily there was a button I could hold down for 15 seconds to reset to factory defaults and then re-upload my most recent config.
I created an Ubuntu VM months ago, for the specific purpose of running Pi-Hole. I gave that up pretty quickly for the exact reason Kris describes.
Weirdly, I’ve been pointing my TrueNAS at a Raspberry Pi 5 running Pi-Hole for a couple of months. I hadn’t run into any issues until RC1 yesterday. The Apps service wouldn’t start until I pointed my TrueNAS at Google’s DNS servers. I also had to disable bridge networking to get the NVIDIA drivers installed.
Weirdly, I checked the Pi-Hole query logs and couldn’t find anything being blocked. But I am using Unbound for DNS resolution, so maybe there’s some recursive wankery happening in the background.
