How iblazhko/temporal-processmanager works
Project type and purpose
• This is a .NET-9, Docker-composed, distributed microservices system that uses Temporal.io as the durable execution engine.
• Purpose: implement a long-running shipment process (classification, orchestration, manifestation, document generation, collection booking, carrier integration) with reliability, retries, timeouts, and observability handled by Temporal.
File structure
• Top‐level “src” folder contains five service hosts, each paired with a Temporal worker project:
• ProcessOrchestrator.Host & ProcessOrchestrator.Temporal
• CarrierIntegration.Host & CarrierIntegration.Temporal
• CollectionBooking.Host & CollectionBooking.Temporal
• DocumentsGeneration.Host & DocumentsGeneration.Temporal
• Manifestation.Host & Manifestation.Temporal
• SharedInfrastructure and SharedKernel libraries provide cross‐cutting configuration, DTOs, domain models, and settings.
• docker-compose.yaml & temporal-dynamicconfig define the Temporal server cluster, UI, and Postgres persistence.
README for additional insights
• Five hosts map to five worker services, each listening on a dedicated Temporal task queue.
• ProcessOrchestrator receives incoming HTTP POST /{shipmentId}, then starts the “entry point” workflow.
• The orchestrator workflow calls child workflows in other task queues (manifestation, docs, collection booking, carrier integration).
• Temporal server persists workflow state and history in Postgres and exposes a Web UI & gRPC API.
Diagramming conventions and elements
• Client → ProcessOrchestrator.Host (HTTP POST)
• Host → Temporal Cluster (gRPC Start/Signal).
• Temporal Cluster → Worker (“poll & task delivery”).
• Worker → Temporal Cluster (“task complete”).
• “Client” lane
• “Application Hosts” lane
• “Temporal Infrastructure” lane
• “External Systems” lane
• Flows with brief labels (“StartWorkflow”, “PollTaskQueue”, “ExecuteActivity”, “ReturnResult”).