Grafana (docker) making tons of DNS requests a second for grafana.com, nonstop. How to find the culprit?

Grafana
,
1

See latest Adguard query logs:

NQoXFtL
NQoXFtL650×973 33.2 KB

This has been going on every second of every day for as far back as my Adguard logs go. To put it into perspective, grafana.com queries are 89% of the total queries reaching my DNS. It’s bad enough to the point that logging the sheer volume of requests is actually having an impact on the life of the host’s SSD.

I have no external plugins phoning home besides the default ones (which can’t be removed), my grafana installation only pulls data from a prometheus instance on my local network. I’ve deleted the default dashboard, and I’ve disabled reporting and updates in the config file:

reporting_enabled = false
check_for_updates = false

I’ve tried inspecting tcpdump on the container, but I’ve only got as far as confirming that the container is in fact bombarding Adguard with these excessive requests, but I have no idea on how to go from there to pinpointing exactly what is causing this behavior.

Any help would be appreciated!

2

Did you change TTL for that record?

I would add host record on the container level, so it won’t be using external DNS (fix any syntax issues):

docker run --add-host=grafana.com=127.0.0.1...
3

I did not, although I did think about it. I’d prefer to actually find what is causing the behavior and correcting it, but it is a really effective plan B

4

So what’s TTL on the container level?

Enable debug logging and otel tracing and good luck to find a root cause.

5

worse case, git clone grafana repo and search for grafana.com

hacky but might help