scadabridge_site_connection_up never counted data connections. It counts open
central->site gRPC subscriber streams (Debug View + the site-wide alarm feed),
so it correctly reads 0 on a healthy site nobody is watching — and the name
invited reading it as device health. That misreading has already cost a live
debugging session: register row 48's 'connection_up is 0 on all six site
nodes' was healthy steady state, not the fault it was diagnosed as.
Two halves, per the row:
RENAME. scadabridge.site.connection.up -> scadabridge.site.stream.subscribers.active
(helpers SiteConnectionOpened/Closed -> SiteStreamSubscriberOpened/Closed), and
the instrument gains the unit:"1" the other gauges already carried. Every
reference updated: the gRPC server, both test suites, Component-Communication.md,
components/Commons.md, docker/README.md and docker/seed-live-probe.sh. A test
now pins the new name AND asserts the old one is gone, so it cannot creep back.
NEW. scadabridge.site.data_connections.connected / .total — the first exported
data-connection signal (the state existed on the health report and in the CLI,
but never as a metric). connected < total is the alert condition; neither
number alone separates 'a device is down' from 'nothing is configured', which
is why the denominator ships with the numerator.
Sourcing is deliberate on two counts. It is fed by the DCL at its four state
transitions rather than derived from the health report, so the metric survives
a dead report pipeline — the exact failure the 'metrics stale' flag exists to
catch, and a derived gauge would go blind at the same moment. And it uses a
static registry reached through ScadaBridgeTelemetry (the SiteConnectionOpened
precedent) rather than a DI-injected holder: the DCL actors already take
ISiteHealthCollector through their constructors, and threading a second
dependency through both would ripple into every test that builds them. It also
keeps SiteHealthCollector untouched, which matters right now — another agent is
concurrently editing that file for the tag-resolution path.
Removal is wired too: an undeployed connection leaves the registry, so it stops
counting toward the total instead of latching connected < total on forever.
Both gauges read 0 on the standby (connections open on the active node only) —
healthy, documented, and called out in the docs so it does not become the next
site_connection_up.
No ObservedMeters allowlist change: all three instruments ride the existing
ZB.MOM.WW.ScadaBridge meter. (An unlisted meter exports nothing, silently.)
Tests: Commons 711 passed (9 new, delta-based against a baseline with unique
per-test connection names, so they do not inherit the register row 32
QueueDepthGaugeTests static-state flake); Communication 717; DCL 319.
A site health report showed tagResolutionCounts 0/0 for a connection whose
dataConnectionTagQuality on the SAME report read good:2 — reproduced live on
the docker rig against site-a's rig-probe-1 instance (2 Good OPC UA tags on
"OPC PLC Simulator"), on the ACTIVE node's own report.
Not a standby-shadowing or keying/removal problem, and not a serialization
problem: both dictionaries live on one collector instance on one node, and the
CentralControlDtoMapper round-trip is correct in both directions. The actor's
own derived counts (TotalSubscribedTagCount / ResolvedTagCount) were right the
whole time — GetHealthReport replied 2/2.
Root cause: HandleSubscribeCompleted — the path that registers tags on a first,
healthy subscribe — mutated _instancesByTag and _subscriptionIds but never
pushed to ISiteHealthCollector. The only pushes were BecomeConnected's (fires
before any instance subscribes, so its 0/0 is legitimate), the batch apply
(reconnect re-subscribe / resolution probe) and unsubscribe (teardown). A
connection whose tags all resolve first time reaches none of those, and
ScheduleTagResolutionRetry cancels the retry timer outright when nothing is
unresolved — so the connect-time 0/0 was never superseded and stood forever.
The blind spot was therefore worst on a perfectly healthy connection, which is
exactly when the "device still booting / tag path unresolved" diagnostic should
read clean rather than empty. Any reconnect masked it (the batch path pushes),
which is why it presented only on a clean first deploy.
Fix: push the derived counts at the end of the first-subscribe apply, mirroring
the existing push site in HandleBatchSubscribeCompleted. No semantic change to
the null/no-data convention — the map entry already existed, its values were
simply stale zeros.
Tests: fail-before regression test pins the exact mechanism (NSubstitute records
a single UpdateTagResolution(name, 0, 0) pre-fix while the actor reports 2/2).
DCL 320, HealthMonitoring 97, SiteRuntime 605, Communication 717 all green;
slnx builds clean.
Docs: Component-DataConnectionLayer.md said the counts were "derived at report
time", which implies a pull that cannot go stale and papered over this bug.
Corrected to state they are pushed, name the four push sites, and record why a
healthy connection was the worst case.
Closes arch-review remediation residual #1 (DCL unsubscribe-during-reconnect
count staleness).
DataConnectionActor tracked TotalSubscribedTags/ResolvedTags as two int fields
incremented and decremented at five independent sites. ReSubscribeAll clears the
very maps those decrements key off (_subscriptionIds, _unresolvedTags) while
deliberately preserving _subscriptionsByInstance, so an unsubscribe landing
inside a reconnect window matched NEITHER decrement branch: the total leaked +1
per subscribe/reconnect/unsubscribe churn cycle, permanently and cumulatively.
The 37f13e2e discard gate stopped the orphan-handle half of that race; it could
not stop the counters drifting, because they were state of their own.
Both counts are now DERIVED at report time from the authoritative per-tag
collections, which makes the drift unrepresentable rather than merely guarded:
total = _instancesByTag.Count (the per-tag counted set the residual
called for — distinct tags with at least
one subscribing instance)
resolved = _subscriptionIds.Count (tags for which the adapter holds a handle)
Two semantic corrections fall out of the derivation:
- A tag whose subscribe failed at CONNECTION level now counts toward the total.
It was excluded before, yet the reconnect re-subscribe re-issued it from
_subscriptionsByInstance and booked it as resolved — resolved above total, and
a total driven negative by the eventual unsubscribe.
- _tagSubscriberCount is deleted. It duplicated _instancesByTag exactly, so
HandleUnsubscribe's last-subscriber test is now "did UnindexTag drop the key?"
— still O(1), with no parallel count that can disagree about when a handle is
released. The subscribe-success promotion split (fresh vs. unresolved→resolved)
also goes: it existed only to pick which scalar to bump; set sizes get
DataConnectionLayer-020's double-count cases right for free.
Behavior is otherwise unchanged — same logging, same handle release, same
unresolved-tag probing, same in-flight-unsubscribe discard semantics (the long
comment block there is updated for the mechanics that changed).
Tests: five TagResolutionCounts_* cases in DataConnectionActorBatchTests
covering the churn repro (3 cycles), a shared tag losing one instance mid
reconnect, connection-level failure then recovery, plain subscribe/unsubscribe
cycles, and a completed reconnect re-subscribe. Verified failing against the
pre-fix actor (churn: total 1 not 0; connection-level: total 0 not 1) and
passing after. Full DCL suite 319/319; solution builds with 0 warnings.
Docs: Component-DataConnectionLayer.md health-reporting section describes the
derived counts; residuals register item 1 marked RESOLVED.
Remove project bookkeeping citations from shipped code comments across the
solution: hyphenated task IDs (WP-14, StoreAndForward-025), milestone/task/
issue refs (M3, Task 4, Audit Log #23, #21), Bundle X task-bundle labels,
and C/D/K/S/T phase labels.
Comment text only — no code logic, string/log literals, or XML-doc structure
changed. Genuine descriptions are preserved (only the citation is stripped),
and technical lookalikes are retained (UTF-8, SHA-256, T00:00:00, M365,
UTC-5, pre-C4/pre-C5 schema versions). Flagged by the new CommentChecker
TaskReferenceInComment / TrackingReferenceInComment checks plus targeted
grep passes; full solution builds clean, append-only guard tests pass.
Fixes the 8 findings from the 2026-06-24 re-review (commit c42bb485), with a
regression test per Medium finding:
- DataConnectionLayer-029 (Med): HandleAlarmSubscribeCompleted now mirrors the
tag-path re-check — if a feed is already stored for the source, release the
redundant just-created subscription instead of overwriting + leaking the first
one (the double-subscribe window DCL-023 reopened). +regression test.
- InboundAPI-031 (Med): remove WaitForAttribute's local 5s grace backstop (tighter
than the CommunicationService Ask's timeout+IntegrationTimeout round-trip budget,
so a slow-but-valid timed-out 'false' got cancelled into a 500). Link only the
client-abort + explicit caller tokens; the lower layer owns the backstop. +test.
- SiteRuntime-032 (Med): derive the deployed count from an authoritative set of
deployed config names (HashSet) instead of a map-presence-gated int, so deleting
a DISABLED instance decrements correctly (SiteRuntime-029's gate leaked it).
+deploy->disable->delete regression test.
- StoreAndForward-028 (Med): reset _bufferedCount in StopAsync alongside the
register-guard so a same-instance Stop->Start re-seeds from a clean base (no ~2N
gauge double-count). +restart regression test.
- AuditLog-017 (Low): test the OnIngestAsync scope-resolution guard (actor survives,
replies empty, counts the failure) — no longer unpinned.
- CentralUI-037 / ScriptAnalysis-009 / SiteRuntime-033 (Low): doc-comment + spec
fixes (Database-throws in the inbound sandbox; baseReferences param wording;
native-alarm cap return-to-normal + per-condition NativeAlarmDropped eviction).
Targeted suites green: SiteRuntime 5, StoreAndForward 6, InboundAPI 31,
DataConnectionLayer 10, AuditLog 5, ScriptAnalysis 40, CentralUI ScriptAnalysis 52.
Remediation from the full per-module code review at 4307c381 (findings recorded
separately in code-reviews/).
Highs fixed:
- DeploymentManager-025/SiteRuntime-031: stop broadcasting notification lists + SMTP
configs (incl. credentials) to sites; site purges already-persisted rows on apply
(enforces the central-only delivery design; clears plaintext SMTP creds at rest).
- DataConnectionLayer-023: guard the native-alarm subscribe path against the
mid-flight-unsubscribe adapter-feed leak (mirrors the DCL-021 tag-path fix).
- SiteEventLogging-024: normalize From/To query bounds to UTC (the -016 fix the
audit trail claimed but never committed).
- KpiHistory-001: add an in-flight guard to the recorder sample tick.
- ScriptAnalysis-001: harden the trust analyzer's TPA-absent fallback (resolve
forbidden anchors in the minimal reference set; warn on degraded mode) — anchors
added to validation references only, never the compile gate.
(InboundAPI-026 left to the feat/ipsen-movein effort per owner decision.)
Medium/Low: DM-026 deterministic deploy-status tiebreaker; SR-027/028/029/030
native-alarm leak/phantom-active/delete-during-redeploy fixes; AL-013/014/016;
TE-024 (folder-mutation audit rows now persisted)/025; SF-025 gauge-provider
clear-on-stop; ESG-025/026; SEC-023/024/025; SCA-007/008/009; plus doc/test
accuracy COM-023/024, HOST-025/026, HM-024/025, NS-027/028.
Full-solution build 0 warnings; ~3560 tests across 18 touched suites green.
STATIC tags (no further OnDataChange after advise) depend entirely on the
seed read. Pre-fix HandleSubscribe seeded only on Success && Value != null,
silently dropping a seed that raced the just-created advise (VT_EMPTY) — so a
static tag stayed Uncertain forever while the source read Good. ReSubscribeAll
did no seeding at all, so a static tag could not self-heal across reconnect.
- New SeedTagsAsync helper: per-tag ReadAsync (not a bulk read — some gateways
time out on large batches) with round-based bounded retry
(SeedReadMaxAttempts/SeedReadRetryDelay), logging any tag that never yields a
value (named — previously zero log trace).
- HandleSubscribe seed loop delegates to SeedTagsAsync.
- ReSubscribeAll re-seeds re-advised tags after reconnect via the
generation-guarded TagValueReceived path (fan-out keys off
_subscriptionsByInstance, preserved across reconnect).
Diagnosed live on wonder-app-vd03 2026-06-17 (see scadabridge-dcl-static-tag-false-bad).
Mechanism #2 (single transient-bad push) left as a follow-up.
The MxGateway alarm mapper emits the SnapshotComplete framing sentinel with
empty SourceReference/SourceObjectReference. HandleAlarmTransitionReceived
routed every transition by prefix match against the subscriber's source, so
the empty-ref sentinel ('' .StartsWith("<src>.") == false) was dropped for
any specific source. The NativeAlarmActor buffers snapshot conditions and only
flushes them on SnapshotComplete, so statically-active native alarms delivered
only in the initial snapshot (no later live transition) never surfaced.
Broadcast the SnapshotComplete sentinel to all alarm subscribers (bypassing the
source match + type filter) so each NativeAlarmActor's snapshot swap completes.
Adds a regression test using the real empty-ref sentinel against a specific
(prefix) source.
DataConnectionActor seeded a tag's initial value by Tell-ing TagValueReceived
from HandleSubscribe's background task, which runs BEFORE HandleSubscribeCompleted
registers the instance's tags in _subscriptionsByInstance. HandleTagValueReceived's
fan-out then found no subscriber and dropped the value. A tag that soon gets a
data-change notification recovers, but a STATIC tag (e.g. an idle MES field that
never changes) was left Uncertain forever — the dropped seed was its only value.
Seeds now ride back on SubscribeCompleted and are delivered after registration,
reusing HandleTagValueReceived's generation guard, fan-out and quality accounting.
+1 regression test (DCL026).
conditionFilter was plumbed end-to-end but applied nowhere — a filtered source
silently mirrored all conditions. Define the filter as a comma-separated,
case-insensitive list of condition type names (blank = all); enforce it
authoritatively client-side in DataConnectionActor routing (uniform across OPC UA
+ MxGateway) and, for OPC UA, additionally build a server-side EventFilter
WhereClause as a bandwidth optimization.
The NativeAlarmActor sends SubscribeAlarmsRequest to the DCL manager, but the
manager only routed tag/write/browse messages to the per-connection
DataConnectionActor — alarm subscribe/unsubscribe were unhandled and dead-lettered,
so native alarms never subscribed at runtime. Caught by live T28 deployment.
Mirrors the existing HandleRoute forwarding.
Expanding a Galaxy object in the tag picker hung on "loading…": the browse
reply inlined every child's full attribute set (~152 KB), exceeding Akka's
128 KB remote frame, and remoting silently discarded the oversized reply.
Browse path (DataConnectionLayer):
- RealMxGatewayClient: navigation now uses BrowseChildren(include_attributes=
false) — child objects only — and an object's own attributes load lazily via
DiscoverHierarchy(root, max_depth=0) when it's expanded. Payload drops from
~152 KB/level to a few KB. Seam contract unchanged.
- DataConnectionActor.CapBrowseChildren: protocol-agnostic byte-budget cap
(~100 KB) on every BrowseNodeResult before it crosses the site→central
frame, OR-ing the adapter's own Truncated flag. Byte budget, not a count —
the only bound that holds regardless of NodeId/attribute-name length.
- RealOpcUaClient: requestedMaxReferencesPerNode 1000 → 500 to narrow the
window before the byte budget applies.
- Graceful gRPC Unimplemented handling → NotSupportedException →
BrowseFailureKind.NotBrowsable with an actionable message (older gateway
builds lacking BrowseChildren).
Picker UI (CentralUI):
- NodeBrowserDialog: modal-lg → modal-xl; new scoped .razor.css caps the tree
at 55vh with its own scrollbar so manual entry + Select/Cancel stay visible.
- Protocol-agnostic failure messages (was hardcoded "OPC UA …"); renamed the
leftover opcua-browser-tree class to node-browser-tree.
Tests: new frame-budget cap test + NotSupported=>NotBrowsable mapping test;
DCL suite 88/88. Doc: Component-DataConnectionLayer.md records the lazy
attribute-light browse and the frame-size guard.
Adds a Test Bindings button to the Connection Bindings table on the Configure
Instance page that opens a modal showing the live current value of every bound
attribute. Reuses the routing path that the OPC UA tag browser landed on:
Central: TestBindingsDialog → IBindingTester → CommunicationService
→ ReadTagValuesCommand → SiteEnvelope (Ask)
Site: SiteCommunicationActor → DeploymentManagerActor singleton
→ DataConnectionManagerActor → child DataConnectionActor
→ _adapter.ReadBatchAsync
Split mirrors the browse handler:
• Manager owns ConnectionNotFound (only it sees the per-site connection set).
• Child owns ConnectionNotConnected (pre-call status check, never stash —
read is interactive design-time), Timeout (OperationCanceledException),
ServerError (any other exception). Per-tag failures from ReadBatchAsync
become failure TagReadOutcomes without aborting the batch.
CentralUI:
• IBindingTester / BindingTester — Design-role guard via HasClaim against
JwtTokenService.RoleClaimType (not IsInRole — see c1e16cf), typed
transport-failure translation.
• TestBindingsDialog — ShowAsync(siteId, rows, instanceLabel) method-arg
pattern (no Razor parameter race; see 2c138b6), groups rows by connection
and issues one ReadAsync per connection in parallel, per-row error subline
+ per-connection banner, Refresh button re-issues the reads.
• InstanceConfigure.razor — Test Bindings button next to Save Bindings,
disabled when no testable rows. OPC UA only today (other protocols have
no ReadTagValuesCommand wiring yet).
Tests:
• Commons: ReadTagValuesCommand discovered by ManagementCommandRegistry.
• DataConnectionLayer: unknown connection → ConnectionNotFound,
not-connected adapter → ConnectionNotConnected (ReadBatchAsync NOT called),
success-path mapping (Good/Bad + per-tag error), cancellation → Timeout.
• CentralUI: register IBindingTester (and the previously-missing
IOpcUaBrowseService) on the existing InstanceConfigureAuditDrillinTests
Bunit container so the page renders cleanly with the new dialog.
- BrowseOpcUaNodeCommand: int DataConnectionId -> string ConnectionName
(site DataConnectionManagerActor indexes children by name; CentralUI
already has the connection name in scope via the dropdown — no extra
plumbing across the trust boundary).
- IOpcUaBrowseService / OpcUaBrowseService: parameter renamed accordingly.
- OpcUaBrowserDialog: collapse the duplicate ConnectionName parameters
(display label and routing key are the same string).
- Task 10: DataConnectionManagerActor forwards BrowseOpcUaNodeCommand to
its child by name (owns ConnectionNotFound); DataConnectionActor adds
the receive across all three lifecycle states (Connecting / Connected
/ Reconnecting) and maps adapter outcomes to BrowseFailureKind
(NotBrowsable / ConnectionNotConnected / Timeout / ServerError).
- Task 17: SetFailure in OpcUaBrowserDialog implements the full
BrowseFailureKind switch with friendly UI messages.
- Tests: DataConnectionManagerBrowseHandlerTests covers ConnectionNotFound,
NotBrowsable, success, and ConnectionNotConnectedException paths.
Joseph Doherty