Skip to content

Guided tour

FabrCore 2.0 · Release and package availability

These guides track the current 2.0 source. Stable 2.0.0 publication is pending; package commands show the release target. Until it is published, follow the source quick start or use a matching available prerelease set. Release migration · Runtime modes

MODULE 13 · LESSON 13.9

Add fleet operations and an optional gateway

Add fleet operations and an optional gateway without changing agent code.

Lesson 82 of 86 · FabrCore 2.0

Overview

A gateway and fleet console are operational choices above the open Host. Insights adds its tenant/cluster/environment management experience. Public application routing, cloud control traffic and direct cluster networking remain separate concerns.

A fleet view aggregates several partial views

Each Host reports its own availability and effective/applied configuration. A management request to one responsive Host cannot establish that every silo or environment succeeded. Group results by the intended deployment targets and explicitly include offline or stale entries. Shared observability stores should not be queried repeatedly as though each Host owned an independent copy.

A gateway is another product boundary

A gateway can route public application traffic to an environment, while the Cloud Server handles control traffic. Correct WebSocket upgrades, authentication, origin policy and timeouts must be verified for that gateway implementation. The open Cloud Server protocol does not automatically supply all routing or tenant UI features a commercial console might provide.

Keep fleet control separate from application routing

  1. Select the target environment and inspect online/offline hosts before a fleet operation. Review configuration history and per-host applied status.
  2. If using a gateway, configure its supported target authentication, routing/origin policy, upgrade forwarding and timeouts. Keep backend endpoints protected.
  3. Test one environment before expanding to multiple clusters. Preserve operation/evidence attribution and distinguish product-specific UI from open protocol contracts.

Test one healthy and one unavailable target

  1. Select a staging environment containing known test targets. Submit a bounded operation and inspect each target's receipt/applied status.
  2. Take one target offline and repeat a read-only fleet check. The aggregate must identify incomplete coverage rather than presenting the remaining successes as a complete fleet result.
  3. If using a gateway, send chat and reconnect a socket through it. Confirm the selected environment/principal and compare behavior with the direct Host endpoint.

This verifies fleet reporting and, when present, gateway routing as separate concerns. A green console connection is not an end-to-end application health check.

If the result is different

Do not claim every Insights screen or paid feature is part of the OSS Host. Validate the deployed gateway's actual policy behavior before exposing it to clients.

Go deeper

Explore the related documentation.