AGH stops responding to DNS requests

[nichu42]

Is anyone else having a problem with AdGuard Home suddenly stopping to respond to DNS requests? I can still log in, but it won't resolve anything until I restart the container.

Could be this: https://github.com/AdguardTeam/AdGuardHome/issues/5896

Don't know how to do that debug profiling with Cloudron, though.

[girish]

@nichu42 I haven't faced this with my instace.

It says you can put debug_pprof: true in the config file . You will find this in /app/data/AdGuardHome.yml .

[girish]

https://github.com/AdguardTeam/AdGuardHome/issues/5944 was closed as duplicate.

Also, technically Cloudron package is on v0.108.0-b.38 . This is worth mentioning if you report it.

[nichu42]

@girish Thanks. I should have mentioned that I was able to set the config, but I don't know to complete the step that says 'open http://127.0.0.1:6060/debug/pprof/goroutine?debug=1'

I've tried both my VPS IP and AdGuard Home's subdomain, but I get a time-out in both cases.

[girish]

@nichu42 Good question, maybe it needs to run from the Web terminal ? Can you curl that url and see?

https://stackoverflow.com/questions/13699297/how-to-use-pprof-in-go-program has some info...

[nichu42]

@girish Indeed, curl works. Thanks!
(And I'm somewhat relieved that no debug info is exposed externally)

[nebulon]

I have seen this from time to time as well. An app restart fixes this. So far I wasn't able to reproduce this reliably though for debugging.

[nichu42]

As far as I understood, no one was able to really reproduce this so far. It just happens. It seems the problem was introduced some releases ago. For me, it started this week. I'm eager to help with troubleshooting. If it cannot be resolved, I suppose it's possible to re-start the app every night via cron or similar for the moment?

[girish]

@nichu42 looks like they found a fix upstream.

[nichu42]

@girish Yes, indeed. Let's hope this really is the problem, and that the fix gets rolled out quickly. Thanks for the notification!

[nichu42]

It looks like the issue was fixed with this release: https://forum.cloudron.io/post/70524
@nebulon You can mark this thread as solved. Thank you!