Installing and upgrading WarehousePG Enterprise Manager

Install WarehousePG Enterprise Manager (WEM) to monitor and manage a WarehousePG (WHPG) cluster. See Architecture for an overview of WEM's components.

Choosing a deployment topology

Before you start, decide where ClickHouse and WEM run relative to your WHPG cluster. This choice determines your installation steps and settings.

• WEM and ClickHouse within the WHPG cluster: typically on the standby coordinator, or the coordinator if your cluster doesn't have one. This is the simplest topology for a single WHPG cluster. That node must run RHEL 8 or RHEL 9, WEM doesn't support RHEL 7 even where WHPG 6.x does.
• WEM and ClickHouse on a dedicated host, outside the cluster: keeps observability components isolated from the cluster's own nodes.

ClickHouse doesn't have to run on either host WEM does. You can instead point WEM at an existing ClickHouse instance you already manage elsewhere, as long as it's reachable from every cluster node's OTel Collector or gateway collector. See Installing ClickHouse for details.

Regardless of topology, install the gateway collector only when multiple systems or clusters send observability data to the same ClickHouse instance. A single WHPG cluster sending data directly to ClickHouse doesn't need one.

The installation process consists of the following steps:

1. Verify the prerequisites.
2. Install ClickHouse, and optionally PostgreSQL, the storage backends for historical metrics, log data, and WEM's own application state.
3. Optionally, install the gateway collector, a centralized OpenTelemetry Protocol (OTLP) ingestion endpoint, if multiple systems or clusters send observability data to the same ClickHouse instance.
4. Install and configure WEM on your dedicated host.
5. Install the host agent on every cluster node to enable privileged operations, such as HBA editing, and to collect host metrics and logs.

If you're upgrading an existing installation, see Upgrading.