​
Purpose

The fleet is the pool of voice workers that actually run conversations. Each worker handles one call at a time, so the size of the fleet is the ceiling on how many calls — inbound and outbound combined — the platform can hold at once. The fleet status page is a read-only monitor. It shows how much of that capacity is in use right now, the health of every worker, and how many free call slots remain. You use it to answer one question at a glance: is there room for more calls, and is every worker healthy? Open it at /dashboard/fleet.
This page has no buttons and nothing to edit. You cannot start, stop, or resize workers from here — that is engineering’s job. Everything on the page is a live indicator you read, not a control you operate.
The page refreshes on its own every five seconds, so the numbers you see are close to real time. There is no refresh button and no need for one — leave the page open and it keeps itself current. While it is fetching the first time, you see a small loading spinner in the centre of the page.

​
The capacity bar

At the top of the page, a single capacity bar summarises the whole fleet. It shows the total number of calls active across every worker against the total capacity of the fleet, written as active / total calls — for example 5 / 100 calls. The bar fills from left to right as usage climbs, and its colour is the headline signal:
Bar colourUsageWhat it means
GreenBelow 80%Plenty of headroom. New calls have somewhere to land.
RedAt or above 80%The fleet is filling up. You are close to the ceiling, and new calls risk having no free worker.
Treat the capacity bar as your at-a-glance health check. If it is green, the fleet is comfortable. If it turns red during a busy campaign, that is your cue to watch it — and, if it stays pinned near full, to ease off dialling pressure by lowering a campaign’s Concurrency Limit. See Pacing and retries.

​
Worker cards

Below the capacity bar, a grid of cards shows one card per voice worker. Each card breaks down that single worker’s identity, health, and load.
Card elementWhat it shows
Worker addressThe worker’s network address, shown in a fixed-width font. Long addresses are trimmed with an ellipsis (…). This identifies the worker if you need to quote it when escalating.
Status icon and labelA health indicator — a green tick reading Healthy, or a red cross reading Unreachable. See Worker status below.
Capacity lineThe worker’s total active calls and how many call slots it provides.
Available lineHow many free call slots the worker has left right now.
Per-worker progress barA small bar showing how full that one worker is, mirroring its share of the capacity.

​
Worker status

The icon and label on each card tell you whether the platform can reach that worker at all.
StatusIndicatorMeaning
HealthyGreen tickThe worker responded to its health check and is ready to take calls.
UnreachableRed crossThe worker did not respond. It is not taking calls, and its slots do not count toward usable capacity.
A worker’s progress bar also turns red when the worker is unreachable, so an unhealthy worker stands out in the grid even before you read its label.
One or two unreachable workers reduce the fleet’s usable capacity but do not necessarily stop calls — the platform routes around them to healthy workers. If several workers are unreachable at once, or an unreachable worker does not recover within a few minutes, that is an infrastructure problem to escalate, not something you can fix from the console.

​
Available slots

The Available line on each card is colour-coded so you can spot a worker that is nearly full without reading the number:
Available slotsColourMeaning
More than 5GreenThe worker has comfortable headroom.
5 or fewerRedThe worker is almost full. Little room left for new calls on this worker.
This is a per-worker view of the same pressure the capacity bar shows for the fleet as a whole. During heavy dialling it is normal for individual workers to flash red as calls land and finish — what matters is the overall picture across the grid and the capacity bar.

​
How to read the page

1

Glance at the capacity bar

Green and well short of full means the fleet is healthy and has room. Red means you are near the ceiling — keep watching.
2

Scan the worker grid for red

A few green workers running near full during a busy campaign is normal. A red Unreachable status, especially on more than one worker, is the thing to act on.
3

Cross-check against your campaigns

If the fleet is full while campaigns are dialling hard, the limit is working as designed — the dialler will not abandon calls it cannot place. If the fleet shows no usable capacity but no campaigns are running, something is wrong.
The fleet is one of the limits that caps how fast a campaign can dial: the dialler only connects an answered call when a free worker is available. If campaigns feel slow and the fleet sits near full, the fleet — not the carrier or the campaign settings — is your real ceiling. See Pacing and retries.

​
Empty state

If no workers are registered, the grid is replaced by a centred No workers registered message instead of cards.
An empty fleet means there is no capacity to take any call — every inbound and outbound call will fail to find a worker. This is never a normal operating state. If you see it, escalate to engineering immediately.

​
When to escalate

The fleet status page tells you what is wrong, but fixing the fleet itself is engineering’s responsibility. Hand these situations to engineering rather than retrying them yourself:
  • The fleet shows no usable capacity but no campaigns are running and workers should be free.
  • The page shows No workers registered.
  • Several workers are unreachable at once, or one stays unreachable for more than a few minutes.
  • The capacity bar stays pinned at full long after you have paused or stopped dialling.
When you escalate, quote the worker address from the affected card and the time you saw the problem — that is what engineering needs to find the worker quickly.
Before escalating “the fleet is full”, check the campaign monitor first. A full fleet during active, heavy dialling is expected behaviour, not a fault — it simply means your campaigns are using all the capacity you have. Pausing or lowering concurrency on a campaign frees slots without any engineering involvement.

​
Next steps

Pacing and retries

How fleet capacity, carrier limits, and campaign concurrency together decide your real dialling ceiling.

Monitor a campaign

Read live campaign activity and confirm whether a full fleet is expected load or a problem.