AG DNS[Crypt?] has become unavailable w/in the last 2 hours or so (sometime between 12p & 2:15p, UTC)
DNS requests should resolve!
They aren't!
Screenshot: Skipping, since easy test
| Description | Value |
| -------------- | ------------ |
| DNS server in use:| (e.g. Default, 176.103.130.130) Either AG DNS via DNSCrypt UI
| How did you setup DNS configuration:| (System/DNSCrypt app/Router) AG Android
| Device model:| (e.g. Google Pixel) Nextbit → Razer Robin
| Operating system and version:| (e.g. Android 7.1.2) Android 7.1.1
@TPS We are on it, thanks mate
fixed, please check
Thanks, all good. Sorry for delay on confirming.
Sorry for this, monitoring failed us yesterday, we should've fixed it right away
I'm just glad I figured it out. I spent an unfruitful few minutes restarting the phone & poking the router until I remembered the custom DNS settings.
Maybe y'all can buildin to the ß versions some sort of diagnostic notification of when DNS &/or DNSCrypt fails this way (for _any_ DNS setting, not just AG DNS)?
Unfortunately, it would be much more complex with DNSCrypt, at least as long at it would silently fail if the server key/cert pair is not completely regenerated. The check what community has been using works only against expired server certificate (the due expiration is shown). We would think about a possibility of sending the corresponding patch to the upstream to expose both types of handshake misbehavior.
Oh, I was thinking along the lines of something more simple, like checking DNS of a known set of URLs on a heartbeat (like when checking filter updates) after determining that IP-address-only connections to those sites work.
DNSCrypt checking could wait, though I thought there was some sort of diagnostic built in to the spec?