Joseph Doherty
fe48458a41
Review of1622d07f: 1. Perf: every i=2253 refresh ran the whole notifier walk + permission checks + QueueEvent under Lock. Now, with nothing deferred (the common case), the refresh is the SDK-shaped walk off Lock (Lock only to copy the root notifiers), kept only if no deferral episode existed at any moment of it (count 0 before and after, episode unchanged); otherwise the cut path also runs and its later replay supersedes. The cut path takes ACL/role verdicts off Lock, then one short Lock hold: watermark + frozen snapshots of Retain conditions from _alarmConditions (no tree walk) + per-item cut and "replay pending" marker; the replay is queued OFF Lock, one event per hold of the item gate. While pending, a condition event newer than the cut marks its condition so the older frozen entry is dropped rather than following it. A concurrent fast-path refresh cancels a pending cut replay of the same item. Measured, 3000 retained conditions, longest Lock wait seen by another thread during one refresh:1622d07f95-106 ms idle / 102-127 ms deferring; now 0-3 ms idle / 23-32 ms deferring. 2. ACL: a retained condition omitted from the replay for want of a grant is never withheld (RefreshCut.RetainedAtCut); only Retain=true events of a condition omitted because it was NOT retained are. 3. One gate/cut state per item for its lifetime (ConditionalWeakTable keyed by the item): an unsubscribe + resubscribe between refresh and drain keeps the cut. 4. Cuts carry their deferral episode; ClearStaleRefreshCuts runs when the count returns to zero, and a stale-episode cut is inert. 5. Dispose guard for the custom refresh (ObjectDisposedException once disposing; Dispose waits for refreshes in flight). 6. Docs: a refresh waits at most for one in-progress QueueEvent on the item. Tests: resubscribe variants, ACL-grant-lands test, pending-replay probe test, dispose test, 3000-condition Lock-hold measurement with a probe pinning that neither path queues its replay under Lock; stress invariant tightened (never backwards outside a refresh; each refresh ends on the newest delivered state) and tolerant of the SDK's own BadRefreshInProgress overlap refusal. Each new test fails against its targeted mutation.
OtOpcUa documentation
Two tiers of documentation live here:
- Current reference at the top level (
docs/*.md) — describes what's shipped today. Start here for operator + integrator reference. - Implementation history + design notes at
docs/v2/*.md— the authoritative plan + decision log the current reference is built from. Start here when you need the why behind an architectural choice, or when a top-level doc says "see plan.md § X".
The project was originally called LmxOpcUa (a single-driver Galaxy/MXAccess OPC UA server) and has since become OtOpcUa, a multi-driver OPC UA server platform. Any lingering LmxOpcUa-string in a path you see in docs is a deliberate residual (executable name lmxopcua-cli, client PKI folder {LocalAppData}/LmxOpcUaClient/) — fixing those requires migration shims + is tracked as follow-ups.
Platform overview
v2 (2026-05-26): the separate
OtOpcUa.Server+OtOpcUa.Adminservices fused into a single role-gatedOtOpcUa.Hostbinary, joined by an Akka.NET cluster. See v2 design for the architectural decision.
- Core owns shared abstractions (driver capability contracts, scripting, virtual tags, alarm historian).
- Drivers plug in via capability interfaces in
ZB.MOM.WW.OtOpcUa.Core.Abstractions:IDriver,IReadable,IWritable,ITagDiscovery,ISubscribable,IHostConnectivityProbe,IAlarmSource,IPerCallHostResolver. Each driver opts into whichever it supports. - Host (
src/Server/ZB.MOM.WW.OtOpcUa.Host) is the single fused binary (.NET 10, AnyCPU).OTOPCUA_ROLESenv decides what to mount:admin(Blazor + control-plane singletons),driver(OPC UA endpoint + per-node actors), or both. See ServiceHosting.md. - Cluster + ControlPlane + Runtime + AdminUI + Security sit between Core and Host. The cluster glues per-node actors into one logical fleet; the control-plane singletons (deploy coordinator, audit writer, redundancy state) live on the admin role-leader. See Redundancy.md.
- The Galaxy driver still reaches MXAccess via gRPC to a separately-installed mxaccessgw sidecar (sibling repo).
Where to find what
Architecture + data-path reference
| Doc | Covers |
|---|---|
| OpcUaServer.md | Top-level server architecture — Core, driver dispatch, Config DB, generations |
| AddressSpace.md | GenericDriverNodeManager + ITagDiscovery + IAddressSpaceBuilder |
| ReadWriteOperations.md | OPC UA Read/Write → CapabilityInvoker → IReadable/IWritable |
| DriverLifecycle.md | Server-side driver lifecycle + infrastructure contracts (IDriverFactory, IDriverProbe, IDriverHealthPublisher, IDriverConfigEditor, IHistorianDataSource) + the Commons library |
| Subscriptions.md | Monitored items → ISubscribable + per-driver subscription refcount (v1 archive) |
| AlarmTracking.md | IAlarmSource + AlarmSurfaceInvoker + OPC UA alarm conditions — native Galaxy alarms end-to-end (live) |
| AlarmTracking.md | Original alarm-tracking write-up (v1 archive) |
| AlarmHistorian.md | Core.AlarmHistorian store-and-forward sink — LocalDbStoreAndForwardSink, the replicated alarm_sf_events buffer + Primary-only drain, IAlarmHistorianWriter, dead-letter/retry/eviction |
| DataTypeMapping.md | Per-driver DriverAttributeInfo → OPC UA variable types (v1 archive — live mapping is in src/Drivers/ZB.MOM.WW.OtOpcUa.Driver.Galaxy/Browse/DataTypeMap.cs) |
| IncrementalSync.md | Address-space rebuild on redeploy + sp_ComputeGenerationDiff |
| HistoricalDataAccess.md | Historical v1 design: history as a per-driver capability (retired; history is now server-side, see Historian.md) |
| VirtualTags.md | Core.Scripting + Core.VirtualTags — Roslyn script sandbox, engine, dispatch alongside driver tags |
| ScriptEditor.md | Monaco script editor — Roslyn-backed IntelliSense (completions, diagnostics, hover, tag-path) for C# virtual-tag scripts; AdminUI/ScriptAnalysis/ backend |
| ScriptedAlarms.md | Core.ScriptedAlarms — script-predicate IAlarmSource + Part 9 state machine |
One Core subsystem is shipped without a dedicated top-level doc; see the section in the linked doc:
| Project | See |
|---|---|
Analyzers (Roslyn OTOPCUA0001) |
security.md § OTOPCUA0001 Analyzer |
Drivers
| Doc | Covers |
|---|---|
| drivers/README.md | Index of the eight shipped drivers + capability matrix |
| drivers/Galaxy.md | Galaxy driver — in-process gRPC client to the mxaccessgw sidecar |
| v1/drivers/Galaxy-Repository.md | Galaxy-specific discovery via the ZB SQL database (v1 archive — the gateway owns this path now) |
For Modbus / S7 / AB CIP / AB Legacy / TwinCAT / FOCAS / OPC UA Client specifics, see v2/driver-specs.md.
Operational
| Doc | Covers |
|---|---|
| Configuration.md | Live appsettings + environment-variable reference (current state) |
| Configuration.md | appsettings bootstrap + Config DB + Admin UI draft/publish (v1 archive — OTOPCUA_GALAXY_* env vars now live in mxaccessgw config) |
| Uns.md | The global /uns master tree — manage Area/Line/Equipment/Tag/VirtualTag fleet-wide; replaces the per-cluster UNS/Equipment/Tags tabs |
| security.md | Transport security profiles, LDAP auth, ACL trie, role grants, OTOPCUA0001 analyzer |
| Redundancy.md | RedundancyStateActor (cluster-scoped singleton), ServiceLevelCalculator, IRedundancyRoleView, per-cluster meshes, Prometheus metrics |
| Reservations.md | Fleet-wide ZTag / SAPID external-ID reservations — publish-time claim, release flow |
| ServiceHosting.md | Single fused OtOpcUa.Host binary install/uninstall with OTOPCUA_ROLES gating; the historian backend is the external HistorianGateway |
| StatusDashboard.md | Pointer — superseded by v2/admin-ui.md |
Client tooling
| Doc | Covers |
|---|---|
| Client.CLI.md | otopcua-cli — OPC UA command-line client |
| Client.UI.md | Avalonia desktop client |
| DriverClis.md | Driver test-client CLIs — index + shared commands |
| Driver.Modbus.Cli.md | otopcua-modbus-cli — Modbus-TCP |
| Driver.AbCip.Cli.md | otopcua-abcip-cli — ControlLogix / CompactLogix / Micro800 / GuardLogix |
| Driver.AbLegacy.Cli.md | otopcua-ablegacy-cli — SLC / MicroLogix / PLC-5 (PCCC) |
| Driver.S7.Cli.md | otopcua-s7-cli — Siemens S7-300 / S7-400 / S7-1200 / S7-1500 |
| Driver.TwinCAT.Cli.md | otopcua-twincat-cli — Beckhoff TwinCAT 2/3 ADS |
| Driver.FOCAS.Cli.md | otopcua-focas-cli — Fanuc FOCAS/2 CNC |
Requirements
| Doc | Covers |
|---|---|
| reqs/HighLevelReqs.md | HLRs — numbered system-level requirements |
| reqs/OpcUaServerReqs.md | OPC UA server-layer reqs |
| v1/reqs/ServiceHostReqs.md | Per-process hosting reqs (v1 archive — only OtOpcUa server hosting remains in scope post-PR-7.2) |
| reqs/ClientRequirements.md | Client CLI + UI reqs |
| v1/reqs/GalaxyRepositoryReqs.md | Galaxy-scoped repository reqs (v1 archive — owned by mxaccessgw today) |
| v1/reqs/MxAccessClientReqs.md | Galaxy-scoped MXAccess reqs (v1 archive — owned by mxaccessgw today) |
| reqs/StatusDashboardReqs.md | Pointer — superseded by Admin UI |
Implementation history (docs/v2/)
Design decisions + phase plans + execution notes. Load-bearing cross-references from the top-level docs:
- v2/plan.md — authoritative v2 vision doc + numbered decision log (referenced as "decision #N" elsewhere)
- v2/admin-ui.md — Admin UI spec
- v2/acl-design.md — data-plane ACL + permission-trie design (Phase 6.2)
- v2/config-db-schema.md — Config DB schema reference
- v2/driver-specs.md — per-driver addressing + quirks for every shipped protocol
- v2/dev-environment.md — dev-box bootstrap
- v2/test-data-sources.md — integration-test simulator matrix (includes the pinned libplctag
ab_serverversion for AB CIP tests) - v2/multi-host-dispatch.md — per-PLC circuit breakers (Phase 6.1 decision #144)
- v2/v2-release-readiness.md — release-readiness tracker
- v2/phase-7-status.md — Phase 7 reconciliation: what shipped vs. the plan, and the five remaining gaps
- v2/implementation/phase--.md — per-phase execution plans with exit-gate evidence
v1 archive
The v1 in-process MXAccess architecture (Galaxy.Host + Galaxy.Proxy + Galaxy.Shared, .NET 4.8 x86 COM, the OtOpcUaGalaxyHost Windows service) was retired in PR 7.2 (2026-04-30, commit ae7106d). Docs that described that shape are kept under v1/ as historical record — see v1/README.md for the index.