Gap 1: Active health probing verified — 60s recovery after platform restart.
Gap 2: Address-based subscription cleanup — no stale handles.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Tested aaBootstrap kill on windev — three gaps identified:
1. No active health probing (IsConnected stays true on dead connection)
2. Stale SubscriptionManager handles after reconnect cycle
3. AVEVA objects don't auto-start after platform crash (platform behavior)
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- MxStatusMapper: maps all 40+ MxStatusDetail codes, MxStatusCategory,
and MxStatusSource to human-readable names and client messages
- OnDataChange: checks MXSTATUS_PROXY.success and overrides quality with
specific OPC UA code when MxAccess reports a failure (e.g., CommFailure,
ConfigError, WaitingForInitialData)
- OnWriteComplete: uses MxStatusMapper.FormatStatus for structured logging
- Write errors: catches COMException separately with HRESULT in message
- Read errors: distinguishes COM, timeout, and generic failures in logging
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Documents when the full STA+Application.Run() approach is needed
(secured/verified writes), why our first attempt failed, the correct
pattern using Form.BeginInvoke(), and tradeoffs vs fire-and-forget.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
The MxAccess docs explicitly state OnWriteComplete always fires after Write().
The real cause is no Windows message pump in the headless service process to
marshal the COM callback. Fire-and-forget is safe for supervisory writes but
would miss secured/verified write rejections (errors 1012/1013).
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
SubscriptionManager tracked client-to-tag routing but never called
MxAccessClient.SubscribeAsync to create the actual COM subscriptions,
so OnDataChange never fired. Now creates MxAccess subscriptions for
new tags and disposes them when the last client unsubscribes.
All 17 integration tests pass.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
- Replaced STA dispatch thread with Task.Run pattern for COM interop
- Fixed TypedValue oneof tracking with property-level _setCase field
- Added x-api-key DelegatingHandler for gRPC metadata authentication
- Fixed CheckApiKey RPC to validate request body key (not header)
- Integration tests: 15/17 pass (reads, subscribes, API keys, connections)
- 2 write tests pending (OnWriteComplete callback timing issue)
- v2 service deployed on windev port 50100
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Replace JoeAppEngine tags with TestChildObject tags (TestBool, TestInt, TestFloat,
TestDouble, TestString, TestDateTime, and array variants) in Phase 4 and Phase 7
plans. These tags cover all TypedValue oneof cases for comprehensive v2 testing.
Design doc covers architecture, v2 protocol (TypedValue/QualityCode), COM threading
model, session lifecycle, subscription semantics, error model, and guardrails.
Implementation plans are detailed enough for autonomous Claude Code execution.
Verified all dev tooling on windev (Grpc.Tools, protobuf-net.Grpc, Polly v8, xUnit).
Generate high-level requirements and 10 component documents derived from source code
and protocol specs. Uses lmxproxy_updates.md (v2 TypedValue/QualityCode) as the source
of truth, with v1 string-based encoding documented as legacy context.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Split LmxProxy Host and Client into a self-contained solution under lmxproxy/,
ported from the ScadaBridge monorepo with updated namespaces (ZB.MOM.WW.LmxProxy.*).
Client project (.NET 10) inlines Core/DataEngine dependencies and builds clean.
Host project (.NET Fx 4.8) retains ArchestrA.MXAccess for Windows deployment.
Added windev.md documenting the WW_DEV_VM development environment setup.
