Known Issues

Current limitations worth knowing about before a show — not bugs waiting to be filed, things the system does on purpose today that may surprise you if you're not expecting them.

Anyone who can reach the dashboard's address and port has full control — there's no username/password, no per-user permissions, and no audit of who made a given change (Events logs what changed, not who changed it). This is permanent, by design, not a gap being closed — see Home's "Getting started" section for what that means for how you should place the gateway on your network. If you need to restrict who can reroute a live show, that has to happen at the network level (firewall, VPN, which physical machines can even reach port 8080), not inside the dashboard itself.

No connectivity tracking for single-feed destinations

RTMP/RIST/plain-UDP-TS destinations (the ones with a single output, no 3-slot vMix-style shape) don't appear in Diagnostics at all, and don't get a reconnect count on Operate either. If you're routing to one of these, Operate's tally-light border is still your live status indicator; it just won't show up in the Diagnostics table or carry RTT/loss/retransmit numbers.

RTT/loss/retransmits only exist for native SRT

Anything going through the ffmpeg bridge (RTMP push/pull, RIST) or plain UDP-TS shows "—" for RTT/packet loss/retransmits in Diagnostics, never a number. This isn't a gap in what's measured — there's no equivalent protocol-level statistic to read from those paths at all (RTMP/RIST/UDP-TS don't expose the same connection telemetry SRT does), so there's nothing to show. Connectivity (the status dot) and traffic counters (bytes/uptime) still work normally for these.

Reconnect count resets when a source or destination is deleted

Deleting and re-adding a source or destination — even if it lands back on the exact same slot — starts its reconnect count, last-error, and RTT/loss history over at zero/blank. This is intentional: a freshly re-created source shouldn't look like it's inherited a previous, unrelated entry's problems. It means "reconnect count" answers "how many times has this configured entry reconnected," not "how flaky has this physical location historically been" across a delete/re-add.

Events log doesn't survive a restart

The Events tab is a live, in-memory session log — useful for "what just happened," not a permanent audit trail. Restarting the gateway (a deploy, an OS reboot, a crash) clears it. If you need a durable record of what happened during a show, this isn't it today.

No presets scoped to a single destination

Applying a preset resets the entire routing matrix to exactly the saved state — any slot not listed in the preset gets cleared, even ones you didn't mean to touch. There's no "apply this preset to just Main Switcher" option; presets are all-or-nothing across every source and destination.

See Future Plan for what's actually being worked toward — some of the above (persistent event logging, a real topology view) are actively on the roadmap; single-feed destination diagnostics and per-destination preset scoping are tracked but not scheduled (see the Backlog milestone); the rest are deliberate, permanent design decisions, not gaps to be closed.