Direct answer

Run /spark health for a live dashboard or /spark health upload to share a static report. Use /spark health show for a quick console summary.

How to check it

  1. Confirm spark is available on your server and that your account has the necessary permission.
  2. Run /spark health show; for details use /spark health show --memory or /spark health show --network.
  3. For a shareable snapshot, use /spark health upload (not --upload).
  4. Record player count and whether the problem was occurring when you captured the report.

What to change

Health shows TPS, CPU, memory and disk-related signals; it does not identify every slow plugin or code path. When performance drops, start /spark profiler start during the issue and finish with /spark profiler stop, then inspect hotspots before adjusting configuration.

Software and version notes

Command availability depends on spark and your server platform. Consult the official command list for differences on proxy, modded and client installations.

Verify the result

Repeat the same workload after a targeted change; compare health snapshots with TPS/MSPT trends and actual player experience. Avoid treating one idle snapshot as evidence of a fix.

References