Files
cameleer-server/.planning/phases/03-agent-registry-sse-push/03-01-PLAN.md
2026-03-11 18:20:45 +01:00

17 KiB

phase, plan, type, wave, depends_on, files_modified, autonomous, requirements, must_haves
phase plan type wave depends_on files_modified autonomous requirements must_haves
03-agent-registry-sse-push 01 execute 1
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentInfo.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentState.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentCommand.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/CommandStatus.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/CommandType.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentRegistryService.java
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentEventListener.java
cameleer3-server-core/src/test/java/com/cameleer3/server/core/agent/AgentRegistryServiceTest.java
cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/AgentRegistryConfig.java
cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/AgentRegistryBeanConfig.java
cameleer3-server-app/src/main/java/com/cameleer3/server/app/agent/AgentLifecycleMonitor.java
cameleer3-server-app/src/main/java/com/cameleer3/server/app/controller/AgentRegistrationController.java
cameleer3-server-app/src/main/java/com/cameleer3/server/app/Cameleer3ServerApplication.java
cameleer3-server-app/src/main/resources/application.yml
cameleer3-server-app/src/test/java/com/cameleer3/server/app/controller/AgentRegistrationControllerIT.java
true
AGNT-01
AGNT-02
AGNT-03
truths artifacts key_links
Agent can register via POST /api/v1/agents/register with agentId, name, group, version, routeIds, capabilities and receive a response containing SSE endpoint URL and server config
Re-registration with the same agentId resumes existing identity (transitions back to LIVE, updates metadata)
Agent can send heartbeat via POST /api/v1/agents/{id}/heartbeat and receive 200 (or 404 if unknown)
Server transitions agents LIVE->STALE after 90s without heartbeat, STALE->DEAD 5 minutes after staleTransitionTime
Agent list endpoint GET /api/v1/agents returns all agents, filterable by ?status=LIVE|STALE|DEAD
path provides
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentRegistryService.java Agent registration, heartbeat, lifecycle transitions, find/filter
path provides
cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentInfo.java Agent record with id, name, group, version, routeIds, capabilities, state, timestamps
path provides
cameleer3-server-app/src/main/java/com/cameleer3/server/app/controller/AgentRegistrationController.java POST /register, POST /{id}/heartbeat, GET /agents endpoints
path provides
cameleer3-server-app/src/main/java/com/cameleer3/server/app/agent/AgentLifecycleMonitor.java @Scheduled lifecycle transitions LIVE->STALE->DEAD
from to via pattern
AgentRegistrationController AgentRegistryService constructor injection registryService.register|registryService.heartbeat
from to via pattern
AgentLifecycleMonitor AgentRegistryService @Scheduled lifecycle check registry.transitionState
from to via pattern
AgentRegistryBeanConfig AgentRegistryService @Bean factory method new AgentRegistryService
Build the agent registry domain model, registration/heartbeat REST endpoints, and lifecycle monitoring.

Purpose: Agents need to register with the server, send periodic heartbeats, and the server must track their LIVE/STALE/DEAD states. This is the foundation that the SSE push layer (Plan 02) builds on. Output: Core domain types (AgentInfo, AgentState, AgentCommand, CommandStatus, CommandType), AgentRegistryService in core module, registration/heartbeat/list controllers in app module, lifecycle monitor, unit + integration tests.

<execution_context> @C:/Users/Hendrik/.claude/get-shit-done/workflows/execute-plan.md @C:/Users/Hendrik/.claude/get-shit-done/templates/summary.md </execution_context>

@.planning/PROJECT.md @.planning/ROADMAP.md @.planning/STATE.md @.planning/phases/03-agent-registry-sse-push/03-CONTEXT.md @.planning/phases/03-agent-registry-sse-push/03-RESEARCH.md

@cameleer3-server-core/src/main/java/com/cameleer3/server/core/ingestion/IngestionService.java @cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/IngestionBeanConfig.java @cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/IngestionConfig.java @cameleer3-server-app/src/main/java/com/cameleer3/server/app/ingestion/ClickHouseFlushScheduler.java @cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/WebConfig.java @cameleer3-server-app/src/main/java/com/cameleer3/server/app/Cameleer3ServerApplication.java @cameleer3-server-app/src/main/resources/application.yml @cameleer3-server-app/src/test/java/com/cameleer3/server/app/AbstractClickHouseIT.java

Pattern: Core module plain class, app module bean config:

  • IngestionService is a plain Java class (no Spring annotations) in core module
  • IngestionBeanConfig is @Configuration in app module that creates the bean
  • IngestionConfig is @ConfigurationProperties in app module for YAML binding

Pattern: Controller accepts raw String body:

  • Controllers use @RequestBody String body, parse with ObjectMapper
  • Return ResponseEntity with serialized JSON string

Pattern: @Scheduled for periodic tasks:

  • ClickHouseFlushScheduler uses @Scheduled(fixedDelayString = "${ingestion.flush-interval-ms:1000}")
  • @EnableScheduling already on Cameleer3ServerApplication

Pattern: @EnableConfigurationProperties registration:

  • Cameleer3ServerApplication has @EnableConfigurationProperties(IngestionConfig.class)
  • New config classes must be added to this annotation

Pattern: ProtocolVersionInterceptor:

  • WebConfig registers interceptor for "/api/v1/data/", "/api/v1/agents/"
  • Agent endpoints already covered -- agents must send X-Cameleer-Protocol-Version:1 header
Task 1: Core domain types and AgentRegistryService with unit tests cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentInfo.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentState.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentCommand.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/CommandStatus.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/CommandType.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentRegistryService.java, cameleer3-server-core/src/main/java/com/cameleer3/server/core/agent/AgentEventListener.java, cameleer3-server-core/src/test/java/com/cameleer3/server/core/agent/AgentRegistryServiceTest.java - register: new agent ID creates AgentInfo with state LIVE, returns AgentInfo - register: same agent ID re-registers (updates metadata, transitions to LIVE, updates lastHeartbeat and registeredAt) - heartbeat: known agent updates lastHeartbeat and transitions STALE back to LIVE, returns true - heartbeat: unknown agent returns false - lifecycle: LIVE agent with lastHeartbeat > staleThresholdMs transitions to STALE (staleTransitionTime recorded) - lifecycle: STALE agent where now - staleTransitionTime > deadThresholdMs transitions to DEAD - lifecycle: DEAD agent remains DEAD (no auto-purge) - findAll: returns all agents regardless of state - findByState: filters agents by AgentState - findById: returns null for unknown ID - addCommand: creates AgentCommand with PENDING status, returns command ID - acknowledgeCommand: transitions command from PENDING/DELIVERED to ACKNOWLEDGED - expireCommands: removes commands older than expiryMs with PENDING status - findPendingCommands: returns PENDING commands for given agentId Create the agent domain model in the core module (package com.cameleer3.server.core.agent):
1. **AgentState enum**: LIVE, STALE, DEAD

2. **CommandType enum**: CONFIG_UPDATE, DEEP_TRACE, REPLAY

3. **CommandStatus enum**: PENDING, DELIVERED, ACKNOWLEDGED, EXPIRED

4. **AgentInfo**: Mutable class (not record -- needs state transitions) with fields:
   - id (String), name (String), group (String), version (String)
   - routeIds (List<String>), capabilities (Map<String, Object>)
   - state (AgentState), registeredAt (Instant), lastHeartbeat (Instant)
   - staleTransitionTime (Instant, nullable -- set when transitioning to STALE)
   - Use synchronized methods or volatile fields for thread safety since ConcurrentHashMap only protects the map, not the values.
   - Actually, prefer immutable-style: store as records in the ConcurrentHashMap and use computeIfPresent to atomically swap. AgentInfo can be a record with wither-style methods (withState, withLastHeartbeat, etc.).

5. **AgentCommand**: Record with fields: id (String, UUID), type (CommandType), payload (String -- raw JSON), targetAgentId (String), createdAt (Instant), status (CommandStatus). Provide withStatus method.

6. **AgentEventListener**: Interface with methods `onCommandReady(String agentId, AgentCommand command)` -- this allows the SSE layer (Plan 02) to be notified when a command is added. The core module defines the interface; the app module implements it.

7. **AgentRegistryService**: Plain class (no Spring annotations), constructor takes staleThresholdMs (long), deadThresholdMs (long), commandExpiryMs (long). Uses ConcurrentHashMap<String, AgentInfo> for agents and ConcurrentHashMap<String, List<AgentCommand>> (or ConcurrentHashMap<String, ConcurrentLinkedQueue<AgentCommand>>) for pending commands per agent.

   Methods:
   - `register(String id, String name, String group, String version, List<String> routeIds, Map<String, Object> capabilities)` -> AgentInfo
   - `heartbeat(String id)` -> boolean
   - `transitionState(String id, AgentState newState)` -> void (used by lifecycle monitor)
   - `checkLifecycle()` -> void (iterates all agents, applies LIVE->STALE and STALE->DEAD based on thresholds)
   - `findById(String id)` -> AgentInfo (nullable)
   - `findAll()` -> List<AgentInfo>
   - `findByState(AgentState state)` -> List<AgentInfo>
   - `addCommand(String agentId, CommandType type, String payload)` -> AgentCommand (creates with PENDING, calls eventListener.onCommandReady if set)
   - `acknowledgeCommand(String agentId, String commandId)` -> boolean
   - `findPendingCommands(String agentId)` -> List<AgentCommand>
   - `markDelivered(String agentId, String commandId)` -> void
   - `expireOldCommands()` -> void (sweep commands older than commandExpiryMs)
   - `setEventListener(AgentEventListener listener)` -> void (optional, for SSE integration)

Write tests FIRST (RED), then implement (GREEN). Test class: AgentRegistryServiceTest.
mvn test -pl cameleer3-server-core -Dtest=AgentRegistryServiceTest All unit tests pass: registration (new + re-register), heartbeat (known + unknown), lifecycle transitions (LIVE->STALE->DEAD, heartbeat revives STALE), findAll/findByState/findById, command add/acknowledge/expire. AgentEventListener interface defined. Task 2: Registration/heartbeat/list controllers, config, lifecycle monitor, integration tests cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/AgentRegistryConfig.java, cameleer3-server-app/src/main/java/com/cameleer3/server/app/config/AgentRegistryBeanConfig.java, cameleer3-server-app/src/main/java/com/cameleer3/server/app/agent/AgentLifecycleMonitor.java, cameleer3-server-app/src/main/java/com/cameleer3/server/app/controller/AgentRegistrationController.java, cameleer3-server-app/src/main/java/com/cameleer3/server/app/Cameleer3ServerApplication.java, cameleer3-server-app/src/main/resources/application.yml, cameleer3-server-app/src/test/java/com/cameleer3/server/app/controller/AgentRegistrationControllerIT.java Wire the agent registry into the Spring Boot app and create REST endpoints:
1. **AgentRegistryConfig** (@ConfigurationProperties prefix "agent-registry"):
   - heartbeatIntervalMs (long, default 30000)
   - staleThresholdMs (long, default 90000)
   - deadThresholdMs (long, default 300000) -- this is 5 minutes from staleTransitionTime, NOT from lastHeartbeat
   - pingIntervalMs (long, default 15000)
   - commandExpiryMs (long, default 60000)
   - lifecycleCheckIntervalMs (long, default 10000)
   Follow IngestionConfig pattern: plain class with getters/setters.

2. **AgentRegistryBeanConfig** (@Configuration):
   - @Bean AgentRegistryService: `new AgentRegistryService(config.getStaleThresholdMs(), config.getDeadThresholdMs(), config.getCommandExpiryMs())`
   Follow IngestionBeanConfig pattern.

3. **Update Cameleer3ServerApplication**: Add AgentRegistryConfig.class to @EnableConfigurationProperties.

4. **Update application.yml**: Add agent-registry section with all defaults (see RESEARCH.md code example). Also add `spring.mvc.async.request-timeout: -1` for SSE support (Plan 02 needs it, but set it now).

5. **AgentLifecycleMonitor** (@Component):
   - Inject AgentRegistryService
   - @Scheduled(fixedDelayString = "${agent-registry.lifecycle-check-interval-ms:10000}") calls registryService.checkLifecycle() and registryService.expireOldCommands()
   - Follow ClickHouseFlushScheduler pattern but simpler (no SmartLifecycle needed -- agent state is ephemeral)

6. **AgentRegistrationController** (@RestController, @RequestMapping("/api/v1/agents")):
   - Inject AgentRegistryService, ObjectMapper
   - `POST /register`: Accept raw String body, parse JSON with ObjectMapper. Extract: agentId (required), name (required), group (default "default"), version, routeIds (default empty list), capabilities (default empty map). Call registryService.register(). Build response JSON: { agentId, sseEndpoint: "/api/v1/agents/{agentId}/events", heartbeatIntervalMs: from config, serverPublicKey: null (Phase 4 placeholder) }. Return 200.
   - `POST /{id}/heartbeat`: Call registryService.heartbeat(id). Return 200 if true, 404 if false.
   - `GET /`: Accept optional @RequestParam status. If status provided, parse to AgentState and call findByState. Otherwise call findAll. Serialize with ObjectMapper, return 200. Handle invalid status with 400.
   - Add @Tag(name = "Agent Management") and @Operation annotations for OpenAPI.

7. **AgentRegistrationControllerIT** (extends AbstractClickHouseIT):
   - Test register new agent: POST /api/v1/agents/register with valid payload, assert 200, response contains agentId and sseEndpoint
   - Test re-register same agent: register twice with same ID, assert second returns 200, state is LIVE
   - Test heartbeat known agent: register then heartbeat, assert 200
   - Test heartbeat unknown agent: heartbeat without register, assert 404
   - Test list all agents: register 2 agents, GET /api/v1/agents, assert both returned
   - Test list by status filter: register agent, GET /api/v1/agents?status=LIVE, assert filtered correctly
   - Test invalid status filter: GET /api/v1/agents?status=INVALID, assert 400
   - All requests must include X-Cameleer-Protocol-Version:1 header (ProtocolVersionInterceptor covers /api/v1/agents/**)
   - Use TestRestTemplate (already available from AbstractClickHouseIT's @SpringBootTest)
mvn test -pl cameleer3-server-core,cameleer3-server-app -Dtest="Agent*" POST /register returns 200 with agentId + sseEndpoint + heartbeatIntervalMs. POST /{id}/heartbeat returns 200 for known agents, 404 for unknown. GET /agents returns all agents with optional ?status= filter. AgentLifecycleMonitor runs on schedule. All integration tests pass. mvn clean verify passes. mvn clean verify -- full suite green (existing Phase 1+2 tests still pass, new agent tests pass)

<success_criteria>

  • Agent registration flow works end-to-end via REST
  • Heartbeat updates agent state correctly
  • Lifecycle monitor transitions LIVE->STALE->DEAD based on configured thresholds
  • Agent list endpoint with optional status filter returns correct results
  • All 7+ integration tests pass
  • Existing test suite unbroken </success_criteria>
After completion, create `.planning/phases/03-agent-registry-sse-push/03-01-SUMMARY.md`