Troubleshooting
A menu option from docs/01 does not exist on my TV
Menu names vary between VIDAA versions and brands. Open the privacy section and read down it: you are looking for anything about viewing information, usage data, content recognition or advertising. Note the real names you find (docs/05 tells you what to record) so the docs can be corrected.
Settings came back after an update
Known VIDAA behavior: some updates reset privacy options or re-ask for consent in a way that re-enables them if you accept quickly. Re-check the privacy section after every update. Thirty seconds.
An app stopped loading after enabling the blocklist
A domain the app needs is probably blocked. Work it as an experiment:
- Open the query log (route 3) and reproduce the problem.
- Look at what was blocked in that moment:
classify_dns_log.pymarksseen_filtered=yesfor exactly those domains. - Unblock the smallest candidate (one domain, or the one functional group you added last) and retest. Keep the smallest unblock that fixes the app.
- Log what you unblocked and why in TESTING.md, so the next firmware update does not send you back to zero.
Known case - Prime Video loses thumbnails and parts of its UI: a fork of the
upstream list reports that four recommendation API hosts
(content-recommendation, api-recommendation, api-recommendation-eu and
api-recommendation-em, all .hismarttv.com) and two recommendation-UI
hosts (detail-ui-eu and recommend-ui-eu, both .vidaahub.com) cause it,
and comments out exactly those six rules to restore Prime Video while
keeping the launcher-level recommendation rules active
(https://github.com/Owen3H/Hisense-VIDAA-DNS-Clean-TV-Blocklist). If you hit
this, comment out only the rules you can confirm, not the whole list. If a rule in the upstream list is the cause, report it there -
breakage reports are how these lists improve.
Icons or artwork load slowly, or appear as empty placeholders
Known side effect reported by the upstream blocklist author: the launcher waits for a blocked resource and then falls back to a cache. It usually settles in a few seconds. If it does not, disable the functional group that causes it.
Requests increase after I unblock
A possible explanation is that a TV retains events locally while DNS fails and sends them when it can resolve again. This is a hypothesis, not a verified behavior of the target TV; firmware behavior is unknown. A burst of requests alone does not establish what data was sent.
fetch_query_log.py: “Cannot reach …”
Run it on the same network as AdGuard Home, and check that the web UI opens
at the same host and port you passed with --host.
fetch_query_log.py: “Redirect … refused”
Give the exact address AdGuard Home listens on (right port, http vs https). The script refuses redirects on purpose so your admin credentials cannot be forwarded somewhere else.
fetch_query_log.py: HTTP 401 or 403
Wrong admin username or password. The password only ever comes from the
hidden prompt or ADGUARD_PASSWORD for that one command - never from the
command line.
AdGuard Home: port 53 is already in use
Another program holds the DNS port. Find it on the AdGuard Home device:
sudo ss -tulpn | grep ':53 '
If it is systemd-resolved (common on Ubuntu and Debian), turn off its stub
listener and point /etc/resolv.conf at the full resolver file:
sudo mkdir -p /etc/systemd/resolved.conf.d
printf '[Resolve]\nDNS=127.0.0.1\nDNSStubListener=no\n' | sudo tee /etc/systemd/resolved.conf.d/adguardhome.conf
sudo mv /etc/resolv.conf /etc/resolv.conf.backup
sudo ln -s /run/systemd/resolve/resolv.conf /etc/resolv.conf
sudo systemctl reload-or-restart systemd-resolved
Run the ss check again; it should print nothing. Then start AdGuard Home.
The device itself cannot resolve names until AdGuard Home is running, because
it now asks 127.0.0.1. To undo, delete the .conf file, restore
/etc/resolv.conf.backup and restart systemd-resolved.
Other common holders: dnsmasq (check it is not serving DHCP before
stopping it), pihole-FTL (run one DNS filter, not two), docker-proxy
(another container maps port 53; see docker ps) and named/bind9.
AdGuard Home works on its own device but other devices time out
nslookup example.com <adguard-ip> from another device times out while
nslookup example.com 127.0.0.1 on the AdGuard Home device answers. Check
which address it listens on:
sudo ss -tulpn | grep ':53 '
If it shows only 127.0.0.1:53, AdGuard Home accepts queries from itself
only. Stop it, then in AdGuardHome.yaml (/opt/AdGuardHome/ for a native
install, the mounted conf directory for Docker) set:
dns:
bind_hosts:
- 0.0.0.0
Keep the indentation as spaces, start AdGuard Home again and re-run the check:
it should show 0.0.0.0:53 or *:53. Stopping first matters, because a
running instance can overwrite the file.
If it already listens on 0.0.0.0 and other devices still time out, look
for a firewall on the device (sudo ufw status, sudo firewall-cmd --state),
a laptop on a guest network or VPN, or Wi-Fi client isolation on the router.
The query log shows no traffic from the TV
The TV is not using the filter. First confirm another device on the same network gets answers from AdGuard Home (see the previous section). Then:
- Check the DNS address on the TV’s network information screen. On some sets DNS can only be changed after switching the whole IP setting to manual; copy the IP address, subnet mask and gateway it showed before.
- Power-cycle the TV at the wall for 30 seconds. Standby can keep the old DNS settings.
- With AdGuard Home in Docker on a bridge network, queries can appear under
a
172.17.x.xaddress instead of the TV’s. Clear the log filter, or run the container with--network host. - If the router announces an IPv6 DNS server, the TV may use that instead. Point it at AdGuard Home or stop announcing it.
- Or check your per-client rules in AdGuard Home / Pi-hole.
In the 2026-11-08 session in TESTING.md, the TV appeared in the log after the listen-address fix above (a TV power cycle was suggested at that point; whether it was done was not recorded).