Nexus Telemetry Fleet documentation

Nexus Telemetry Fleet monitors and manages a fleet of Starlink terminals from one dashboard. A telemetry collector beside each terminal reports to a fleet server, managed or self-hosted. Start with the quick start, or open a section below.

Start here

Create an account, get a fleet server, and put the first Starlink terminal on the dashboard. The quick start is the whole path in 5 steps.

Quick start: The requirements and the 5 steps from sign-up to a terminal on the dashboard.
Account and fleet server: Create your account in the console, add a fleet server, choose its address, and sign in for the first time.
Add a collector: Generate an enrolment token, install the collector on the host, and the terminal joins the dashboard.
Dashboard: The Status view lists the terminals that need attention first. The 5 states, the other views, the live view and search.

Telemetry collector

Install the Starlink telemetry collector on Docker, Debian, Red Hat, macOS or Windows, keep it on a release channel, and retire it when a site closes.

Docker: Pull and verify the image, enrol the container, run it, update it and remove it.
Debian: Install the collector from the signed apt repository, enrol it, run it, update it and remove it.
Red Hat: Install the collector from the signed rpm repository, enrol it, run it, update it and remove it.
macOS: Install the signed and notarised package on a Mac, enrol from the setup page or the command line, and update or purge it.
Windows: Install the collector from the MSI, enrol it from the setup page or PowerShell, and update or purge the host.
Release channels: The 3 channels, where a collector's channel is recorded, updates from the dashboard, and turning them off.
Setup page: The page a collector serves on its own host, for enrolment and then for its status.
Onboard status page: The page a collector serves on the site network that shows whether a connection problem is on their own device, the site network or the Starlink connection.
Collector GPS: The GPS source for a collector, set on the terminal's Admin view or in the collector's configuration file.
Retire and revoke: Decommission, revoke, remove, re-admit and move a collector, and reset one on its host.

Fleet dashboard

Monitor the whole Starlink fleet: the status views, attention cards, groups, alerts, maintenance windows, enrolment tokens, operators, the availability report and the audit trail.

Views and filters: The 5 views of the dashboard, the rail, the top bar, the filters and the search palette.
Attention cards: The cards under Needs attention on the Status view, incidents, and the terminal panel a card opens.
States: The 5 dashboard states, the terminal's own state words, freshness, and how a fault is confirmed.
Groups: Create a group, add terminals to it, rename or delete it, and filter the dashboard by group.
Tokens: The outstanding enrolment tokens, tokens that admit many collectors, and revoking a token.
Certificates page: The tiles, panes and audit view of the Certificates page, and the Rotate action.
Users and roles: The 4 roles, and adding a user, limiting a user to groups, resetting a password or a second factor, and deleting a user.
Sign-in: Passwords, the email address on an account, two-factor authentication, the sign-in policy, and Microsoft sign-in.
Alerts: The 4 alert triggers, the email and webhook channels, test alerts and the delivery log.
Organisations: The organisation tree, what a reseller and a customer each see, and suspending or cancelling an organisation.
Maintenance windows: Maintenance windows, acknowledgements and silences: the 3 ways to stop alerts about a known problem.
Availability report: The monthly availability report: what it counts as up, down, unobserved and planned, and its export.
Audit trail: The record of every administrative change and who made it, for one organisation and for the whole fleet server.
Server health: The page that reports on the fleet server itself: its checks, its capacity, and what degraded means.
Settings: The account, organisation and fleet server settings, how each saves, and testing a connectivity target.

Starlink terminal

Manage one Starlink terminal: its live telemetry, sky map, alerts, reachability diagnostics and settings.

Live view: The live telemetry of one terminal: its state, its instruments, its charts, and the speed test and reboot actions.
Obstruction: The sky map of one terminal: where its view of the sky is blocked, and the signal, alignment and boot panels.
Alerts: The alerts of one terminal: what is open now, its past outages, and its event history.
Diagnostics: The reachability of one terminal from its collector to the internet, and what the terminal reports about itself.
Terminal settings: The Admin view of one terminal: the collector's configuration, and the actions that manage the collector.

Self-hosted server

Run the fleet server on your own hardware: install with Docker Compose or a package, harden it, back it up, update it and license it.

Install: Run a fleet server on hardware you own, from the compose files or the package, and create its first user.
Hardening: TLS for the dashboard, the health check addresses, logs and metrics, rate limits, and opening the dashboard to other hosts.
Backup and restore: What a backup contains, taking one, copying it off the host on a schedule, and restoring it.
Server updates: Updating a self-hosted fleet server by package or by Docker, rolling back, and the update notification.
Licensing: Activating a self-hosted fleet server with a licence key, the collector ceiling, and what happens after the key expires.

Reference

Every collector and server setting, environment variable, command, timing, alert rule, file location and log event.

Collector configuration: Every key in collector.toml and every NF_ variable the collector reads, with its default.
Server configuration: Every NF_ variable the fleet server reads and every fleet server field on the Settings page, with defaults and ranges.
Command line: Every nf-server and nf-collector command, its flags, and the exit codes.
Timings: Every state and freshness word, and every timing, retention period and certificate lifetime, with where each is set.
Alert rules and webhook: The alert triggers, the repeat-suppression rules, and the payload of every webhook post.
Files and repositories: Where the collector and the server put their files, and the repositories and images they come from.
Roles and permissions: One matrix of who can see and do what, by role.
Log events: Every named event the fleet server and the collector write to their logs, when it fires, and what to do.

Trust

How certificates and the root key secure the fleet, what data leaves a site, and the answers a security review asks for.

Certificates: The 4 things a collector's trust rests on, how renewal happens without anyone, and the one task left to a user.
Root key: What the root key is and is not, where it is kept, and what the Root key card on the Certificates page shows.
Best practice: Backing up the root key, extending the signing cover, taking the key offline, restoring it, and a quarterly check.
Data sent: What the collector sends to the fleet server, what never leaves the site, and how long it is kept.
Security FAQ: What connects where, how organisations are kept apart, who holds the keys, what a compromise exposes, and how to leave.

Help

Troubleshooting by symptom, how to reach support, and the known limitations.

Troubleshooting: The common faults, listed by what the dashboard or the log shows, with the steps for each.
Support: How to contact support during the beta, and what to include so the problem can be found.
Known limitations: What Nexus Telemetry Fleet does not do during the beta, and what to do about each.

New to the subject? The Starlink fleet monitoring guide explains what the terminals report and why a collector beside each one matters.