Deployment Guide
This guide takes you from evaluating Mattermost through to running it in production: planning a deployment, preparing its dependencies, installing the server, securing and scaling it, and rolling out the desktop and mobile apps.
Start here
Work through the sections below in order. Each one links to the pages inside it.
Two situations call for a different starting point. To evaluate Mattermost and see its capabilities before committing to a deployment plan, begin with Quick Start Evaluation. If your environment has particular constraints, such as no internet egress, a coalition network, or an intermittently connected edge, read Deployment Scenarios first, because those constraints shape every decision that follows.
Deployment scenarios
Deployment Scenarios describes the operational patterns Mattermost is deployed into - out-of-band communications, mission partner collaboration, DDIL operations, sovereign collaboration in Microsoft, and air-gapped environments. Read this before the sections below if any of them describe your environment, because the constraints shape every decision that follows.
Server deployment
The steps below run in order. Each is a section in the sidebar.
- Plan - Understand the components of a deployment, size it for your user count, decide whether you need high availability and clustering, and check the software and hardware requirements.
- Prepare - Stand up the prerequisites before you install: the database, file storage, network access, an NGINX reverse proxy, and TLS. Large deployments also provision search infrastructure and caching with Redis here.
- Choose a deployment method - Compare the methods, then install on Linux, Kubernetes, or with containers.
- Secure your deployment - Encryption at rest and in transit.
- Back up and recover - Backup strategy and active/passive disaster recovery on AWS.
- Air-Gapped Operations - The runbook, bill of materials, registry mirroring, and configuration changes for a deployment with no internet access.
- General deployment troubleshooting - Log review, environment review, and the diagnostic data to gather for a support ticket. It also indexes the guides for specific layers - Docker, the database, and the desktop and mobile apps.
Calls deployment
Calls Deployment - Deploy and operate Calls infrastructure, including RTCD, the recording offloader, logging, and metrics.
Deploy the apps
- Desktop App Deployment - Distribution and management for Windows, macOS, and Linux, including MSI and group policy.
- Mobile App Deployment - EMM distribution, Microsoft Intune MAM, custom builds, and running your own push proxy.