Skip to main content
Version: 2026.07.11

Restart a Brain Service

Canonical source: 00_MISSION_CONTROL/runbooks/brain-services/RUNBOOK_Brain_Service_Restart.md

:::danger Use the service-ops tool Live Brain currently runs as SYSTEM scheduled tasks. Use the service-ops tool so restart, port cleanup, and health checks stay consistent. :::

One command

pwsh -NoProfile -File B:\GRIFF_WARM\AI_AGENT_PLATFORM\00_MISSION_CONTROL\service-ops\Restart-GriffService.ps1 -Action restart -Service all
GoalCommand
StatusRestart-GriffService.ps1 -Action status -Service all
Restart MCPRestart-GriffService.ps1 -Action restart -Service Brain-MCP
Restart HTTPRestart-GriffService.ps1 -Action restart -Service Brain-HTTP
JSONadd -Json

Healthy output shows manager=schtask, taskActionOk=True, and healthy=True.

Live tasks

Logical serviceScheduled taskPort
Brain HTTP\GRIFF-Brain-HTTP8787
Brain MCP\GRIFF-Brain-MCP8788

The tool ends the task, clears orphaned port holders, starts it again, checks health, and writes a receipt under 06_SESSION_EVIDENCE/receipts/.

True checks

Invoke-RestMethod http://127.0.0.1:8787/ready
$headers = @{ Authorization = "Bearer $env:GRIFFAI_MEMORY_API_TOKEN" }
Invoke-RestMethod http://127.0.0.1:8787/l1l5/status -Headers $headers

Expected: /health returns version 1.2.0.dev7, a nonempty release build and the router/arbiter grounding contract; /ready returns the current live item count and 166 schema objects. Do not hard-code corpus counts. After a restart, run the governed A+ gate; 28/28 is the current accepted result.

Do not disturb

Restarting Brain does not authorize broad migrations, VACUUM, WAL checkpoint, Qdrant image cutover or changing ingest/OMNI policy. The event-driven ingest watcher is the live primary lane and its CUDA embedding work is separate from a Brain service restart.