Files

T

Joseph Doherty 07a4a5ff5e docs(mqtt): P1 plain-MQTT milestone live-verified; endpoint recorded

Task 14 — the P1 milestone gate. Ran the live /run verification on a docker-dev
rig against the real Mosquitto TLS+auth fixture, and recorded the result.

The gate did its job: it found MQTT was unauthorable in production.

  RawDriverTypeDialog's driver-type list is a hand-maintained array that nobody
  added MQTT to, so the /raw "New driver" picker never offered it — with every
  other layer (contracts, driver, browser, factory, host registration, typed tag
  editor) complete and 266 green unit tests. Nothing tied that array to
  DriverTypeNames and AdminUI has no bUnit, so no offline test could see it.
  Fixed, plus RawDriverTypeDialogParityTests: a reflection parity guard that
  fails for ANY DriverTypeNames entry missing from the picker. Verified
  falsifiable — RED with "…cannot be authored from the /raw New-driver picker,
  so they are unreachable in production: Mqtt" before the one-line fix.

A second gap is left OPEN and documented, not fixed (no task ever covered it):
MqttDriverForm / MqttDeviceForm do not exist, and DriverConfigModal/DeviceModal
have no raw-JSON fallback — so an operator can create an MQTT driver but cannot
author its broker endpoint or credentials. P1 is not operator-complete until
those two razor forms land. The gate authored config via SQL to get past it.

Live gate result (isolated 2-node MAIN stack, image built from this branch,
fixture at 10.100.0.35:8883, AllowUntrustedServerCertificate rather than
pinning the CA into the rig):

  - typed MQTT tag editor renders and round-trips; Json shows the JSON-path
    field, Raw/Scalar hide it; wildcard topic a/+/c blocked at save; data-type
    list offers Float64 and no Double; qos/retainSeed absent on "(driver
    default)" and qos:2 written as a number; isHistorized/historianTagname
    survive a topic-only edit; SparkplugB renders the P2 placeholder and
    switching back keeps Plain values; jsonPath is guidance not a gate in all
    three shapes ($ seeded only for a brand-new blank tag, blank stays absent,
    clearing saves fine)
  - deployed, and both editor-written blobs resolved and served changing live
    values through OPC UA (22.8 / 1629, StatusCode Good) — no BadNodeIdUnknown
  - the bespoke #-observation browser drove a real broker session and rendered
    the fixture's actual topic tree (line1/{counter,state,temperature},
    noise/chatter, retained/seed); a Modbus device's picker still correctly
    reports "Browsing unavailable"

Also found live and documented (not a code change): both nodes of a redundant
pair run the driver, so a FIXED clientId makes them evict each other forever —
the broker logs "already connected, closing old connection" and both nodes
reconnect every ~2s while still reporting Healthy. Unset (the default) is
correct; confirmed 0 reconnects/60s on both nodes after removing it.

Suites: offline 1134 passed / 0 failed (266 Driver.Mqtt + 730 AdminUI incl. the
3 new guards + 138 Core.Abstractions); live 7/7 against the fixture.

Docs: new docs/drivers/Mqtt.md + Mqtt-Test-Fixture.md (the per-driver + fixture
convention every other driver follows), MQTT rows in docs/drivers/README.md,
TestConnectProbes.md, root README.md, and infra/README.md §3.

CLAUDE.md drift corrected while adding the MQTT endpoints:
  - the "every fixture carries project: lmxopcua" claim was false — only the
    MQTT fixture does; the filter returns one stack, not the fleet
  - lmxopcua-fix.ps1 is Windows-VM-only; on macOS drive the host over ssh/rsync
    (and the docker-dev rig itself runs locally under OrbStack)

Claude-Session: https://claude.ai/code/session_01GASWkNEi68FSCtvr6rLoEW

2026-07-24 18:44:43 -04:00

12 KiB

Raw Blame History

Infrastructure — testing & deployment

Canonical index of every piece of infrastructure OtOpcUa's tests and deployments depend on: the shared Docker host, the driver integration fixtures, the env-gated live-gate recipes, the integration-test harness, the local dev rig, and the gateway/deploy setups. This file is the single entry point; deeper history lives in the linked docs.

Credentials below are dev-only (per decision #137) — never production values. Production uses Integrated Security / gMSA (SQL) and a real directory (LDAP).

Related deep docs

docs/v2/dev-environment.md — full historical inventory + Windows bootstrap.
docs/v2/Galaxy.ParityRig.md — mxaccessgw / Galaxy parity-rig runbook.
docs/drivers/*-Test-Fixture.md — per-driver fixture coverage maps.
../scadaproj/infra/glauth/ — the shared GLAuth source-of-truth + deploy runbook.
reference_histgw_local_deploy, reference_ww_historian_vm (memory) — HistorianGateway deploy detail.

1. Shared Linux Docker host — `10.100.0.35`

All Docker workloads (central SQL, driver fixtures, integration-harness deps) run on a shared Debian host, not on the dev machine. Discover everything with the project=lmxopcua label:

ssh dohertj2@10.100.0.35 'docker ps --filter label=project=lmxopcua --format "{{.Names}}\t{{.Status}}"'

From this Mac, drive it over passwordless SSH (ssh dohertj2@10.100.0.35 'docker …'). docker -H ssh://… does not work from the Windows VM (stdio bridge hangs) — the Windows VM uses the lmxopcua-fix.ps1 helper (~/bin) instead; the Mac uses SSH directly.

Stacks live at /opt/otopcua-<driver>/ (root-owned) or ~/otopcua-<driver>/ (user-owned, where /opt needs sudo). Current layout:

Stack dir	Purpose
`/opt/otopcua-mssql`	central SQL (always-on)
`/opt/otopcua-modbus` · `/opt/otopcua-abcip` · `/opt/otopcua-s7` · `/opt/otopcua-opcuaclient` · `/opt/otopcua-mqtt`	driver fixtures
`~/otopcua-ablegacy` · `~/otopcua-focas`	driver fixtures (user-owned)
`~/otopcua-harness`	Host.IntegrationTests real-mode deps (SQL + GLAuth) — see §4

The repo's tests/.../Docker/ trees are the source of truth; the host dirs are mirrored deployments. Push changes with rsync/lmxopcua-fix sync before docker compose up.

2. Central SQL Server (always-on)


Container	`otopcua-mssql` (`mcr.microsoft.com/mssql/server:2022-latest`), stack `/opt/otopcua-mssql`
Endpoint	`10.100.0.35:14330` → `1433`
Creds (dev)	`sa` / `OtOpcUaDev_2026!`
Used by	`appsettings.json` `ConfigDb`; DB-backed test suites (`.Configuration.Tests`, `.Admin.Tests`, `*.Server.Tests`)

Nuclear reset: ssh dohertj2@10.100.0.35 'cd /opt/otopcua-mssql && docker compose down -v && docker compose up -d'.

3. Driver integration fixtures

Each driver's integration suite TCP-probes its fixture at collection init and skips cleanly when it isn't running. Bring one up, dotnet test, tear down. Repo compose lives under tests/Drivers/….IntegrationTests/Docker/.

Driver	Fixture image	Endpoint	Bring up (on the host)
Modbus	`otopcua-pymodbus:3.13.0`	`10.100.0.35:5020`	`docker compose --profile <standard\|dl205\|mitsubishi\|s7_1500> up -d`
AB CIP	`otopcua-ab-server:libplctag-release`	`10.100.0.35:44818`	`docker compose --profile <controllogix\|compactlogix\|micro800\|guardlogix> up -d`
AB Legacy (PCCC)	`otopcua-ab-server` (PCCC mode)	`10.100.0.35:44818`	`AB_LEGACY_COMPOSE_PROFILE=<…> docker compose up -d` (stack `~/otopcua-ablegacy`)
S7	`otopcua-python-snap7`	`10.100.0.35:1102`	`docker compose --profile s7_1500 up -d`
OpcUaClient	`mcr.microsoft.com/iotedge/opc-plc:2.14.10`	`opc.tcp://10.100.0.35:50000`	`docker compose up -d`
FOCAS	`otopcua-focas-sim` (vendored focas-mock)	`10.100.0.35:8193`	`docker compose up -d` (stack `~/otopcua-focas`)
MQTT	`eclipse-mosquitto:2.0.22` (+ a publisher sidecar)	`10.100.0.35:8883` TLS+auth · `:1883` plaintext+auth	`MQTT_FIXTURE_USERNAME=otopcua MQTT_FIXTURE_PASSWORD=<pw> docker compose up -d` (stack `/opt/otopcua-mqtt`)

Endpoint overrides point a suite at a real PLC instead of the sim: MODBUS_SIM_ENDPOINT · AB_SERVER_ENDPOINT · S7_SIM_ENDPOINT · OPCUA_SIM_ENDPOINT · MQTT_FIXTURE_ENDPOINT / MQTT_FIXTURE_PLAIN_ENDPOINT.

MQTT needs credentials + material, unlike the other fixtures. The broker has no anonymous fallback on either listener (deliberate — an anonymous broker would let a bad auth config pass), so it will not even start until ./gen-fixture-material.sh has written secrets/ (password file + CA + server cert/key, gitignored) and MQTT_FIXTURE_USERNAME/MQTT_FIXTURE_PASSWORD are exported at docker compose up. The live suite additionally wants MQTT_FIXTURE_CA_CERT pointing at a local copy of the CA: scp dohertj2@10.100.0.35:/opt/otopcua-mqtt/secrets/ca.crt /tmp/mqtt-fixture-ca.crt. It is also the only fixture whose services carry the project: lmxopcua label (see CLAUDE.md).

Fixture-cycling gotcha: profile-gated services can share a port; a plain docker compose down may leave a stale container bound. Force-remove before switching profiles: docker rm -f $(docker ps -aq --filter name=otopcua-<driver>) then up --force-recreate.

4. Host.IntegrationTests real-mode harness

tests/Server/ZB.MOM.WW.OtOpcUa.Host.IntegrationTests runs on EF-InMemory + a stub LDAP by default (zero infra). Real-mode swaps in SQL Server + a real GLAuth bind. Stack: ~/otopcua-harness (repo: tests/Server/.../Host.IntegrationTests/docker-compose.yml + glauth/config.toml).

Dep	Container	Endpoint	Notes
SQL	`otopcua-harness-sql-1`	`10.100.0.35:14331` → 1433	port 14331 avoids the always-on 14330
GLAuth	`otopcua-harness-ldap-1`	`10.100.0.35:3894` → 3893	GLAuth (swapped in for retired bitnami/openldap, PR #451); users below

Harness GLAuth dev users (glauth/config.toml, baseDN dc=zb,dc=local): serviceaccount/serviceaccount123 (search), alice/alice123 (full-access), bob/bob123 (read-only).

ssh dohertj2@10.100.0.35 'cd ~/otopcua-harness && docker compose up -d'
export OTOPCUA_HARNESS_USE_SQL=1  OTOPCUA_HARNESS_USE_LDAP=1
export OTOPCUA_HARNESS_SQL_HOST=10.100.0.35,14331
export OTOPCUA_HARNESS_LDAP_HOST=10.100.0.35        # harness uses port 3894
dotnet test tests/Server/ZB.MOM.WW.OtOpcUa.Host.IntegrationTests
ssh dohertj2@10.100.0.35 'cd ~/otopcua-harness && docker compose down'

5. Live-gate recipes (env-gated destructive tests)

These prove failure/recovery behavior against real infra and skip cleanly offline. Each is gated by env vars and always restores the shared resource in a finally.

5a. S7 connect-timeout blackhole (`S7_1500ConnectTimeoutOutageTests`)

Proves an S7 subscription survives a SYN-blackhole outage (frozen PLC), which a docker restart can't reproduce. docker pause gives the ACK-less-established shape — no sudo/VPN.

ssh dohertj2@10.100.0.35 'cd /opt/otopcua-s7 && docker compose --profile s7_1500 up -d'
export S7_TIMEOUT_OUTAGE_START_CMD='ssh dohertj2@10.100.0.35 "docker pause otopcua-python-snap7-s7_1500"'
export S7_TIMEOUT_OUTAGE_STOP_CMD='ssh dohertj2@10.100.0.35 "docker unpause otopcua-python-snap7-s7_1500"'
dotnet test tests/Drivers/ZB.MOM.WW.OtOpcUa.Driver.S7.IntegrationTests --filter "FullyQualifiedName~ConnectTimeoutOutage"

5b. LDAP GLAuth outage/recovery (`LdapGlauthOutageLiveTests`)

Drives the directory-outage circuit through the production auth path against the harness GLAuth (§4), paused/unpaused. Also verifies the GLAuth swap binds correctly.

ssh dohertj2@10.100.0.35 'cd ~/otopcua-harness && docker compose up -d ldap'
export GLAUTH_LIVE_HOST=10.100.0.35   # port defaults to 3894 (GLAUTH_LIVE_PORT)
export GLAUTH_OUTAGE_START_CMD='ssh dohertj2@10.100.0.35 "docker pause otopcua-harness-ldap-1"'
export GLAUTH_OUTAGE_STOP_CMD='ssh dohertj2@10.100.0.35 "docker unpause otopcua-harness-ldap-1"'
dotnet test tests/Server/ZB.MOM.WW.OtOpcUa.Host.IntegrationTests --filter "FullyQualifiedName~LdapGlauthOutageLive"

5c. HistorianGateway LiveIntegration soak (`Category=LiveIntegration`) — needs VPN

Exercises the gateway read/write/alarm/EnsureTags paths against a real AVEVA historian. Requires the corporate VPN up and a running gateway (see §6). Full deploy detail: reference_histgw_local_deploy.

export HISTGW_GATEWAY_ENDPOINT=http://localhost:5221   # local gw, h2c
export HISTGW_GATEWAY_ALLOW_UNTRUSTED=true              # required for h2c
export HISTGW_GATEWAY_APIKEY=histgw_placeholder_placeholder   # gw ApiKeys:Mode=Disabled
export HISTGW_TEST_TAG=SysTimeSec
export HISTGW_WRITE_SANDBOX_TAG=HistGW.LiveTest.Sandbox
export HISTGW_BOOL_SANDBOX_TAG=HistGW.LiveTest.BoolSandbox
export HISTGW_ALARM_SOURCE=OtOpcUaLiveValidation
dotnet test tests/Drivers/ZB.MOM.WW.OtOpcUa.Driver.Historian.Gateway.Tests --filter "Category=LiveIntegration"

6. HistorianGateway — deploy for live testing

The gateway sidecar runs locally in Docker (~/Desktop/HistorianGateway/docker/) pointed at the real AVEVA historian over VPN. It's the backend for OtOpcUa's historian read/alarm/continuous paths.

Hosts (VPN up): historian gRPC CAP wonder-sql-vd03.zmr.zimmer.com:32565 (TLS); Runtime SQL :1433 (wonderapp/Wonderapp1, SQL login); Galaxy GR SQL wonder-app-vd03.zmr.zimmer.com:1433.
Deploy: publish FDD on host, then docker compose up -d --build. gRPC h2c localhost:5221, dashboard/health localhost:5220, ApiKeys:Mode=Disabled. Corporate hostnames go in a gitignored docker/.env / secrets, never in committed compose.
Known deploy fixes (committed to the gateway repo): FQDN via .env; libgssapi-krb5-2 gss-ntlmssp in the Dockerfile (historian v6 + SQL paths auth NAM\dohertj2 via Negotiate/GSSAPI); RuntimeDb on.
Full recipe + gotchas: reference_histgw_local_deploy · historian VM: reference_ww_historian_vm.

7. docker-dev local rig — `docker-dev/docker-compose.yml`

A 2-node OtOpcUa cluster built from source for AdminUI / OPC UA /run verification.

Surface	Endpoint	Notes
AdminUI	`http://localhost:9200` (via Traefik)	round-robins central-1/2 — rebuild both to verify a UI change
OPC UA	`opc.tcp://localhost:4840` (central-1) · `:4841` (central-2)	data plane
Login	disabled (`Security__Auth__DisableLogin=true`)	auto-authenticated full-access admin — no sign-in needed
LDAP (data plane)	shared GLAuth `10.100.0.35:3893` (`DevStubMode=false`)	gates Client.CLI read/write role ops only

docker compose -f docker-dev/docker-compose.yml up -d --build central-1 central-2 traefik
# headless deploy: POST http://localhost:9200/api/deployments  (X-Api-Key)
docker compose -f docker-dev/docker-compose.yml down -v   # reset

8. Shared GLAuth (data-plane auth) — `10.100.0.35:3893`

The project-wide LDAP for docker-dev + live OPC UA data-plane auth (distinct from the harness GLAuth on :3894). Container zb-shared-glauth, baseDN dc=zb,dc=local, bind cn=serviceaccount,dc=zb,dc=local/serviceaccount123, test users password password. Source of truth + deploy runbook: ../scadaproj/infra/glauth/.

9. Deployment (production rig)

The deploy host runs the OtOpcUa server + mxaccessgw + the HistorianGateway sidecar as NSSM-wrapped services from C:\publish\. Refresh after landing changes with scripts/install/Refresh-Services.ps1 (stops reverse-dependency order → backs up → publishes → restarts → smoke-verifies ports 4840/4841). Production creds use Integrated Security / gMSA (SQL) and a real directory (LDAP) — never the dev values in this doc.

The legacy Wonderware TCP/ArchestrA sidecar was retired; the historian backend is now the ZB.MOM.WW.HistorianGateway sidecar (see §6 and docs/Historian.md).

12 KiB Raw Blame History