You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Use the docker container to setup a clean, empty install of latest development snapshot (using :nightly), issue is also reproducable with 6.0.4 (:latest). Do not change any settings, do not use pihole for your DNS requests, simply download and set it up. Click around on the admin web quickly. You will see some strange lags. I can reproduce it by switching between Dashboard and Query Log quickly for several times.
I cannot see anything unusual on the locks debug output. Butfor me it looks like locks are not released somewhere on concurrent webserver requests that query the database.
I can reproduce it by switching between Dashboard and Query Log quickly for several times.
It may simply be that you are exhausting the webserver workers as modern browsers are not closing connections they've opened as they don't know if they will be reused later. Try increasing webserver.threads to a significantly larger value (like 200) and see if this mitigates this. If so, then this is actually expected. Fortunately,
I can reproduce it by switching between Dashboard and Query Log quickly for several times.
is not a realistic use case :-)
Anyway, if locks would not be released, everything would freeze as there are no timeouts on locks. You could possibly track down what is happening by setting debug.all = true either via the API, the CLI, env variable or pihole.toml directly.
Use the docker container to setup a clean, empty install of latest development snapshot (using :nightly), issue is also reproducable with 6.0.4 (:latest). Do not change any settings, do not use pihole for your DNS requests, simply download and set it up. Click around on the admin web quickly. You will see some strange lags. I can reproduce it by switching between Dashboard and Query Log quickly for several times.
I cannot see anything unusual on the locks debug output. Butfor me it looks like locks are not released somewhere on concurrent webserver requests that query the database.
This is the docker-compose.yml I use:
The text was updated successfully, but these errors were encountered: