Choose your components
To decide what to install, start with what you want OpenLM Platform to collect. Each component runs in your own environment, on a license server, an end-user machine, a machine that can reach your directory, or a host in your network, and most of them send data to OpenLM Platform. Use this page to see which component does which job, where it runs, how you deploy it, and where its installation guide is.
Choose your components
Each component does one job and runs on its own machine. Most of them connect to OpenLM Platform with an authorization file of their own client type, which you generate in Identity, as described in Authorize components. Download the installers from the OpenLM downloads page. The OpenLM Router is the exception: you pull its container images from the public registry public.ecr.aws/r3q3q2f4, as described in Deploy OpenLM Router. The Cloud Broker needs no installation: you connect it in OpenLM Platform. See Cloud Broker.
The following table compares the components:
| Component | What it does | Install it on | Installers | Connects with | Installation guide |
|---|---|---|---|---|---|
| Broker | Collects license usage from your license managers | Each license server, or a machine next to it | Windows Installer, tar.gz archive, DEB package, RPM package | Authorization file of type Broker | Install Broker |
| Workstation Agent | Collects application usage on end-user machines | Each end-user machine | Windows Installer, tar.gz archive, DEB package, RPM package, macOS package | Authorization file of type Agent | Install Workstation Agent |
| Directory Synchronization Agent (DSA) | Collects users and groups from your directories | A Windows machine that can reach the directory | Windows Installer | Authorization file of type Dsa | Install the Directory Synchronization Agent (DSA) |
| AI Proxy | Collects AI usage, for AI FinOps reports | A Docker host in your network | Docker Compose | The Location of your OpenLM Platform, plus a Client ID and Client Secret on a secured OpenLM Platform, set in the AI Proxy UI | Deploy AI Proxy, Troubleshoot AI Proxy |
| SaaS Agent | Collects licensing data from your SaaS accounts | A Windows machine | Windows Installer | Authorization file of type SaaSAgent | Install SaaS Agent |
| OpenLM Router | Moves FlexLM clients from a failed license server to its backup | A 64-bit Linux host with Docker, in your network | Container images, run with Docker Compose | Authorization file of type Router, imported in the Router UI | Deploy OpenLM Router, Configure OpenLM Router |
For a new OpenLM Platform account, follow Get started to install the Broker, then the Workstation Agent, then the DSA.
Install and configure OpenLM Broker
The Broker collects license usage from the license managers on a license server, and sends it to OpenLM Platform through the Broker Hub. Install one Broker on each license server, or on a machine next to it. On Windows, the installer includes a Java runtime. On Linux and Unix, install a Java runtime first.
After the installation, approve the Broker in the Broker Hub, and then connect its license managers and approve them. These guides cover each step:
- Install Broker on Windows: the installer and silent installation.
- Install Broker on Linux or Unix: the tar.gz archive, the DEB and RPM packages, and
detect.sh. - Change, upgrade, or uninstall Broker: TLS certificates, upgrades, and removal.
If the Broker connects to a self-hosted OpenLM Platform whose certificate the Broker machine doesn't trust, import that certificate as described in that guide.
For an interactive walkthrough of the Windows installation, step through this demo:
Install and configure Workstation Agent
The Workstation Agent runs on end-user machines. It reports which licensed applications each user runs and for how long, and sends the data to OpenLM Platform through the Agents Hub. Install it on each end-user machine. The agent then runs for each user who signs in.
After the installation, check the agent in Agent Activity. These guides cover each platform:
- Install Workstation Agent on Windows: the installer, silent installation, and deployment to many machines with tools such as Microsoft Intune, Configuration Manager, or Group Policy.
- Install Workstation Agent on Linux: the DEB package, the RPM package, and the tar.gz archive.
- Install Workstation Agent on macOS: the package, on one Mac or across a fleet with mobile device management (MDM).
- Change, upgrade, or uninstall Workstation Agent: settings, a new authorization file, upgrades, and removal.
For a self-hosted OpenLM Platform, keep OpenLM On-Premise in the Windows installer and enter the address of your Agents Hub, as described in Connect to a self-hosted OpenLM Platform.
For an interactive walkthrough of the Windows installation, step through this demo:
Install and configure Directory Synchronization Agent (DSA)
The Directory Synchronization Agent (DSA) reads users and groups from your directories, local (Lightweight Directory Access Protocol, LDAP) or in the cloud (Microsoft Entra ID, Google, Okta), and sends them to OpenLM Platform. Install it on a 64-bit Windows machine that can reach the directory.
After the installation, approve the agent in Directory Sync, and then add your directory and define what to sync. For the installer, silent installation, approval, and troubleshooting, see Install the Directory Synchronization Agent (DSA).
For an interactive walkthrough of the installation, step through this demo:
Install and configure AI Proxy
You deploy AI Proxy as a Docker Compose stack in your network, not through a desktop installer. These guides cover each step:
- AI Proxy overview: what it does and how it fits OpenLM Platform.
- AI Proxy system requirements: the host, the hardware, and the network.
- Deploy AI Proxy: run the Docker Compose stack in your network.
- Configure AI Proxy: connect it to OpenLM Platform with the Location and, on a secured OpenLM Platform, a Client ID and Client Secret, set up providers, and route clients.
- Troubleshoot AI Proxy: fix a stack that doesn't start, connect, or report usage.
Install and configure SaaS Agent
The SaaS Agent is a Windows application that collects licensing data from your SaaS accounts. Its guides cover the installation, the connection to OpenLM Platform, and day-to-day operation:
- Install SaaS Agent: what it collects, and the host and Windows account to prepare.
- Install the SaaS Agent: the installer, or a silent installation.
- Connect the agent to OpenLM: generate the authorization file in Identity and import it.
- Change, upgrade, or uninstall SaaS Agent: run, pause, and upgrade the agent, or remove it.
- Troubleshoot SaaS Agent: find the logs, and fix an agent that doesn't install, start, connect, or scan.
Deploy OpenLM Router
OpenLM Router moves FlexLM clients from a failed license server to its backup on its own. The clients point at the Router instead of at a license server, so nothing changes on their machines during a failover. The Router runs as a Docker Compose stack on a 64-bit Linux host in your network. It needs the Router feature in your OpenLM license, and the Broker on each license server in a failover pair, so that OpenLM Platform has the health of each server.
After the deployment, connect the Router to OpenLM Platform with an authorization file of type Router, and then define your failover pairs in OpenLM Platform. These guides cover each step:
- OpenLM Router: how failover works, and what you deploy.
- System requirements: the license, the host, the datastores, and the network.
- Deploy OpenLM Router: the images, the Compose stack, and upgrades.
- Configure OpenLM Router: the connection to OpenLM Platform, failover pairs, and the client settings.
- Change, upgrade, or remove OpenLM Router: upgrades, a new host, and removal.
- Troubleshoot OpenLM Router: fix a Router that doesn't connect or fail over.
Generate the authorization file of the first component you chose, as described in Authorize components.
The documentation team might draft this page using a large language model. A technical writer reviews every page and verifies it against the product before publishing it. A human makes the final decision.