Feedback
Platform Health
This dashboard shows heartbeat (check-in) information for the Relativity Processing queue manager and worker hosts. It shows only the running heartbeat; other Invariant health checks (for example RabbitMQ, website, certificate, database, worker restart rate, and process failures) are not displayed on this dashboard. It does not report resource usage, job health, or error conditions. It lists every queue manager and worker host that has reported a running heartbeat, together with the agent version, how long the process has been running, when it started, and when it last checked in. It helps administrators confirm that processing servers are checking in, running the expected version, and have not recently restarted.
Availability
The Platform Health dashboard is available in Relativity Server 2026 and later. It depends on the running health check instrumentation shipped with the Invariant processing components. If the tables are empty, the environment does not include the instrumentation yet, or no heartbeats were received in the selected time range.
Prerequisites
- Environment Watch is installed and the Elastic/Kibana stack is receiving telemetry from the target Relativity Server.
- The Relativity Server is version 2026+ so that Invariant queue manager and worker health checks are emitted.
- Queue manager and worker processes have been running within the selected time range so that heartbeats are available.
Navigation
The Platform Health dashboard is part of the Processing feature domain. A tab strip at the top lets you switch between:
- Processing Overview
- Platform Health
- Inventory & Discovery
A left-hand navigation panel provides quick links to other Environment Watch feature domains.
Filters and Time Range
- Time range - Use the time picker in the top-right to choose the window. Only hosts that sent a heartbeat inside the selected range appear in the tables, so a host that stopped reporting before the range began will not be listed.
- KQL filter bar - Use the query bar to narrow results, for example by host name.
Server Health Tables
The dashboard contains two tables with the same columns:
- Worker Manager Servers - Queue manager hosts.
- Worker Servers - Worker hosts.
| Column | Description |
|---|---|
| Host | Name of the host that sent the heartbeat. |
| Agent Version | Version of the processing process running on the host. |
| Running For | How long the process has been running since its last start, shown as a readable duration (for example, 3 days or 10 hours). |
| Start time | Time the process started. |
| Last Check-In | Time of the most recent heartbeat received from the host. |
Heartbeats are sent every 10 to 15 minutes, so a Last Check-In within that window is normal. A Last Check-In that is far older than the heartbeat interval suggests the process is stopped or the host is unreachable. A recent Start time or small Running For value on a host that should be stable indicates a restart. Differing Agent Version values across hosts point to a partial or incomplete upgrade.
Metrics Reference
Log sources
Log name (labels.name) |
Description |
|---|---|
relsvr.invariant.queuemanager.runninghealthcheck
|
Running health check heartbeat from a queue manager host. |
relsvr.invariant.worker.runninghealthcheck
|
Running health check heartbeat from a worker host. |
Fields
| Field | Description |
|---|---|
labels.host_name
|
Host that emitted the heartbeat. |
labels.process_version
|
Version of the reporting process. |
numeric_labels.process_uptime_days
|
Days since the process started. |
invariant_process_start_time_display
|
Process start time. |
invariant_last_check_in_display
|
Time of the last check-in. |
Panel-to-metric mapping
| Panel | Type | Log name | Key fields |
|---|---|---|---|
| Worker Manager Servers | Table | relsvr.invariant.queuemanager.runninghealthcheck
|
labels.host_name, labels.process_version, numeric_labels.process_uptime_days, invariant_process_start_time_display, invariant_last_check_in_display |
| Worker Servers | Table | relsvr.invariant.worker.runninghealthcheck
|
labels.host_name, labels.process_version, numeric_labels.process_uptime_days, invariant_process_start_time_display, invariant_last_check_in_display |
Use Cases
| Use Case | Description |
|---|---|
| Verify server availability | Confirm every expected queue manager and worker host is listed with a recent last check-in. |
| Detect unexpected restarts | Look for hosts with a short running time or recent start time. |
| Confirm upgrade consistency | Compare agent versions across hosts after an installation or upgrade. |
| Troubleshoot stalled processing | When jobs are not progressing, check whether workers or queue managers have stopped reporting. |
Associated Alerts
No processing-specific Environment Watch alerts are currently associated with this dashboard. This section will be updated as processing alerts are published.
