Before assuming a feature is broken, check whether the bot itself is actually the problem: a live Resources page shows RAM and CPU usage, database and websocket latency, and which infrastructure node is currently running the bot for that server, updated roughly every couple of seconds. Most "the bot isn't responding" reports are a missing setting, not an outage — but when it genuinely is an outage or lag spike, this page is where to confirm it instead of guessing.
What the Resources page actually shows
- Live usage: RAM used versus total, heap usage and CPU load, with a small scrolling graph of the last 60 seconds so a spike is visible rather than just a single current number.
- Uptime and node: how long the current process has been running, and the name of the specific node serving your server — relevant because Noxyr runs on more than one node, and a custom bot in particular runs on whichever node it was assigned to.
- Live latency panel: separate round-trip numbers for the database, Redis cache, the Discord websocket connection, message delay and node-to-node latency, each colored green, amber or red depending on how high it currently is.
- Service status: online/starting/offline for the main bot process, MongoDB and Redis specifically — the three things that have to all be healthy for every feature to work.
The page updates itself over a live connection when your browser supports it, falling back to polling every couple of seconds otherwise — there is nothing to refresh manually.
Reading it: what is normal versus a real problem
Latency figures in the green range across database, websocket and message delay, with RAM under the point where the meter turns amber, is a normal healthy state — expect small fluctuations, not a flat line. A single latency number briefly spiking amber is not worth acting on; a service showing "Offline" for more than a few seconds, or every latency figure red at once, is worth waiting a minute and checking again before assuming your specific server's setup is at fault.
When to check this page versus checking your own settings
- One feature not working, everything else fine — this is almost always a configuration issue (a missing channel, a role hierarchy problem, a module switched off), not an infrastructure one. Check the Overview page's setup checklist first.
- Nothing is responding at all, across every feature — check Resources before anything else. If the main bot service shows offline or every latency number is red, the fix is on the infrastructure side and no dashboard setting will change that.
- Slow, not broken — a command takes a few extra seconds but eventually works. Check the message delay and database latency figures specifically; a temporarily elevated number that recovers within a minute or two is a blip, not an outage.
Why the node matters, especially for a Custom Bot
Noxyr runs its bot processes across more than one node for capacity. For most servers this is invisible — you never need to know or care which one you are on. It becomes relevant for Custom Bot specifically, since a custom bot is assigned to a particular node when it starts, and the Resources page is where you can confirm which one and see that node's own numbers rather than a global average. If a custom bot is behaving oddly, the node name on this page is useful to mention if you contact support, since it points at exactly which server-side process to check.
What this page cannot tell you
It reports infrastructure health, not feature configuration. A server with a perfectly healthy bot and green latency across the board can still have a ticket panel that never posts because the category was never set, or a welcome message that never fires because the toggle is off. Resources answers "is the bot itself okay," never "is my server's setup correct" — for that, the Overview page's setup checklist and each feature's own dashboard page are the right place to look.
Common questions
Does high RAM usage mean the bot is about to crash? Not on its own — the meter turns amber and then red as thresholds are crossed specifically so a sustained climb is visible before it becomes a problem, rather than being a guaranteed sign of an imminent crash.
Why does my server's node name matter if I only run one server? For a normal (non-custom) server it mostly does not — it is informational. It matters more once you run a Custom Bot, since that bot's own node is what actually determines its latency and uptime independent of the main bot's health.
Is this the same as the /status command? No — /status reports your community's own numbers (members, tickets, sales). Resources reports the bot's infrastructure health, which is a completely different kind of "status."
Can members see this page? No, it is part of the admin dashboard for a server and requires the same access as any other dashboard page — server owner or Administrator.
What should I do if Resources shows everything healthy but my feature still is not working? Treat it as a configuration problem, not an infrastructure one. Check the Overview page's setup checklist and the specific feature's settings page for a missing channel, role or toggle.
Check your server's live status on the Resources page of your dashboard, and see the dashboard docs for what else lives alongside it.