Skip to Content
This documentation is provided with the HEAT environment and is relevant for this HEAT instance only.
InternalCluster ManagerPlatform service logs (Cluster Manager)

Platform service logs (internal)

Engineering one-pager for HEAT platform log surfaces and the Cluster Manager collector.

Surfaces

ServiceMechanismEndpoint
core-apiHEAT.Common HEATLogger in-memory ledgerGET /_logs/coreapi, GET /_logs/coreapi/pretty
heat-authSameGET /_logs/auth, GET /_logs/auth/pretty
heat-v2-apiSerilog console onlyNo ledger; Cluster Manager uses kubectl logs deployment/heat-v2-api

Ledgers default to 1000 entries per pod (HEATLogger.cs). Entries go to ledger + stdout; auditor ingest is explicit (HeatAuditorClient.cs), not automatic.

Cluster Manager

  • Collector: serviceLogCollector.ts
  • Routes: GET /api/environment/service-logs, GET /api/environment/service-logs/:serviceId (admin)
  • Wired in app.ts via startServiceLogCollector()
  • Advisories read getServiceLogSnapshots() from AdvisoryContext

Lovable / UI authoring pack

Full operator-oriented one-pager (field glossary, triage rules, examples):

references/lovable/cluster-manager/16-platform-service-logs.md

Env vars

VariableDefault
SERVICE_LOG_COLLECTOR_ENABLEDtrue
SERVICE_LOG_COLLECTOR_INTERVAL_MS60000
SERVICE_LOG_LOOKBACK_MS900000
SERVICE_LOG_MAX_CACHED_ENTRIES500
SERVICE_LOG_KUBECTL_SINCE5m
SERVICE_LOG_KUBECTL_TAIL500
CORE_API_URLcluster: http://core-api; local dev: http://localhost:5220
HEAT_AUTH_URLcluster: http://heat-auth; local dev: http://localhost:5168
HEAT_NAMESPACEheat