The dashboard

The banner that matters most

If anything is saved and not applied, there is a yellow banner at the top saying so, with the version nginx is actually running. That is the single most useful thing on the page, because "I changed it and it did not work" is nearly always "you saved it and did not apply it".

The counts

Sites, backend pools, certificates and fleet size, each with a one line summary underneath: how many are enabled, how many pools are fully healthy, whether every certificate is current, how many nodes are online. Anything that is not in a good state says so rather than showing a number you have to interpret.

What nginx is doing

One row per node: active connections, requests per second, connections per second, dropped, reading, writing and waiting. These come from stub_status, sampled every fifteen seconds, with rates worked out between two samples.

Dropped is the one to look at. It is connections nginx accepted and could not handle, and it is nearly always zero. When it is not, something is wrong and it is worth finding out what.

Pool health

Every pool with its balancing method, how many backends are up out of how many, and which sites use it. A pool that is partly down is visible here before anybody reports a slow site.

This box

The nginx version, the OpenSSL version with a post quantum badge when the build supports it, how many modules are compiled in, whether stream proxying works, and the current config version. It is the quickest way to confirm two nodes are running the same thing.

The top right corner

Which node you are looking at and whether it is active or standby. This matters more than it looks. Both nodes show the same screens, but changes are made on the active one. If you find a button grayed out, check here first.

Common questions