Re-review 1: HandleDispatch's terminal guard and its AwaitingApplyAcks flip read the row separately, so a
lost-dispatch closure committed between them was silently reopened. The flip now carries the RowVersion the
guard read; a change in between is a concurrency conflict (nothing seeded, nothing published, coordinator
reset). Pinned on real SQL Server with a closure injected between the two reads (fails without the fix).
AdminOps' apply deadline and the coordinator's now both read DeployTimings.ApplyDeadline, passed explicitly
to both singletons. Microsoft.IdentityModel.Protocols(.OpenIdConnect) pinned to 8.11.0 and referenced
directly by the Security project, so the whole IdentityModel family resolves 8.11.0 (fresh restore clean
of NU190x/NU1605).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Ruling refinement on I1. Only timeouts are strikes, so a never-returning
while(true), which times out once and is then only refused, was never
quarantined, and that defeated E1. A refusal now quarantines immediately
when the orphan it was refused behind has been executing for longer than
ScriptingOptions.StuckOrphanQuarantineFactor x the script's timeout
(default 10, so 20 s at the 2 s default; 0 or less disables it). It uses
the same quarantine path: one Error, otopcua.script.quarantined with
reason=stuck, Bad, lifted on source change or redeploy. A young orphan
(a short overrun) is still a plain refusal.
The guard stamps the run's start (Stopwatch timestamp) when the pool
picks it up and clears it when the run really finishes.
Tests: the 30-trigger while(true) regression (paced) now ends
quarantined with reason "stuck", still with exactly one orphan and a
healthy sibling. Positive control: a 2x overrun is refused, not
quarantined, and evaluates again once the orphan completes. A factor of
0 disables the rule. Docs (ScriptEditor "Runtime safety", VirtualTags)
and the seam remarks are updated.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
BoundedApiKeyStore kept the set (volatile) and its DateTimeOffset stamp (non-volatile, 16 bytes) in separate
fields, so a reader could pair a new set with an old stamp or read a torn stamp; they are now one immutable
snapshot behind a volatile reference. An inner store that cannot list ids now switches the set off with a
per-instance flag instead of writing KnownIdSet = false into the shared ApiKeyStoreGuardOptions.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
RefreshAsync now tells "the store answered: no such secret" (revocation - drop the refreshed key AND the
lazily cached one) apart from "the read failed" (keep the previous key). Store reads fan out through
Parallel.ForEachAsync capped at MaxConcurrentRefreshReads = 8 instead of one read per cluster at once.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The one-snapshot refactor re-sorted every collection in memory with an ordinal comparer, while the old
composer's SQL ORDER BY used the database's case-insensitive collation; mixed-case keys therefore re-hashed
an unchanged configuration (spurious drift fleet-wide, a "new" revision on a no-op redeploy). The snapshot
factory now reads every composed table ORDER BY its natural key server-side and Compose serialises in the
order given. Pinned byte-for-byte against a verbatim replica of the old composer on the real SQL Server.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
GalaxyTelemetryTests (ActivityListener) and GatewayGalaxyDataWriterFailClosedTests
(MeterListener, filtered only by meter and instrument name) observe every Galaxy span or
measurement in the process. They now run in a collection with parallelization disabled
(F-FU6), because a parallel driver test, or its background probe seed, polluted their counts.
The stream-cancel assertion in GalaxyDriverShutdownTests goes back to a bounded wait: the
instantaneous IsCompleted check introduced in round 1 was racy.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Review fix round 2 for lane F:
- N1: the init platform seed no longer runs inline under the init token. Galaxy's Tier-A
Initialize budget is 5 s, and a cold gateway DiscoverHierarchy can spend all of it, so the
OperationCanceledException rethrow failed init, opened the Initialize breaker and cost a full
rebuild per retry. InitializeAsync now only resolves the platform source (no network) and
queues a background seed tied to a per-runtime CTS that the teardown cancels. A result that
lands after its runtime was torn down is discarded.
- N2: the seed worker serves the runtime that is live when each pass STARTS, so a request made
under a new runtime while a stale pass is in flight always gets a pass of its own.
- N3: SubscribeAsync and the subscribe-once read capture the runtime generation before
SubscribeBulk and register only if it is unchanged, atomically with the teardown's registry
Clear (_registryGate). On a mismatch they release the handles and fail the call. Re-subscribe
Rebind uses the same gate, and the production rebuild clears the registry unconditionally.
- PerPlatformProbeWatcher re-checks its stop after the SubscribeBulk await, atomically with
the record and publish. A pass that straddles StopAsync releases the handles it acquired
instead of writing the probe entry back into a cleared registry.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
I7: a Dispatching row committed after the coordinator's PreStart whose Tell never arrived had no owner and
refused every later deploy. The in-flight check (now fully off the admin singleton's thread, serialised by the
actor-local compose flag alone) closes a Dispatching row with zero NodeDeploymentState rows that is past the
apply deadline as TimedOut (optimistic on RowVersion) and proceeds.
HandleDispatch ignores a dispatch for a Sealed/PartiallyFailed/TimedOut row, so a late or duplicated dispatch
cannot flip it back. Coordinator status/reason writes attach a key+RowVersion stub instead of loading the
multi-MB ArtifactBlob; proven on the real SQL Server (blob untouched, seal lands).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
SQL Server's DbUpdateException for one duplicate names every entry in the statement batch, so detaching
ex.Entries dropped the whole flush. On a row-level rejection the writer now asks the database which of the
batch's EventIds it already holds, drops exactly those, re-saves the rest, and falls back to row-by-row only
if a non-duplicate row is still refused. Fresh context + entities per attempt. Pinned against the real
central SQL Server (new SqlServerTestDatabase harness) and a faithful whole-batch interceptor.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- ScriptLogRateLimiter.LimitPerWindow. The ScriptLogTopicSink and
VirtualTagActor summaries now print the configured limit, not
DefaultLimitPerWindow. Tests assert "limit N/window" for a non-default
N.
- CalculationDriver.Dispose cancels, then takes _evalGate (bounded 2 s)
and disposes the gate and _evalCts. The pump and timer ticks enter the
gate through TryEnterEvalGateAsync, which stops cleanly once the
driver is disposed. Test: Dispose with a 6 s runaway run in flight
returns promptly, disposes both, and a late trigger is harmless.
- ScriptedAlarmEngine captures _shutdownToken once and disposes
_shutdownCts in Dispose when every drained task has completed. A task
stuck past the grace leaves the CTS to the GC. Lifecycle tests assert
the disposal.
- DraftValidator: CalculationNoEffectiveTrigger is skipped for a
change-triggered tag whose scriptId does not resolve. That case is
already CalculationScriptMissing or CalculationScriptNotFound, and
there are no reads to count. Both existing tests now assert the
absence (red before the change).
- docs/VirtualTags.md: System.Threading is denied as a whole namespace,
with an allow-list; the type list is now accurate (Unsafe,
RuntimeHelpers, and the ForbiddenMembers on System.Type); the resource
limits section covers the run guard and the allocation budget.
docs/ScriptEditor.md: accurate deny-list and a new "Runtime safety"
section covering timeout, refusal, quarantine after 3 timeouts,
allocation budget, lifting, and rate-limit summaries that appear only
on the source's next entry after a burst then quiet.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The window used to count every terminal row, so 50+ failed deploys pushed the last good artifact out of it.
Now the newest KeepSealed (>= 1) Sealed rows keep their blob, failed rows are prunable unless referenced in
the ReferencedDays window. Pinned: 60 TimedOut + 1 old Sealed -> Sealed retained, 60 TimedOut emptied.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The fixture wedged EVERY write on one shared token-ignoring release gate, so
the later same-tag write was also hung, and the 300 ms pessimistic bound
abandoned it while the test was still polling before ReleaseWrites(). Its
reason was "write did not complete within" (the bound), not "queued for" (the
queue budget), and it had already entered the driver, so the tag had been
released. The release path is correct. Only the first write now hangs, and
later writes complete at once.
The abandon reason formatted the bound with TotalSeconds:F0, which printed a
300 ms bound as "0s". It now uses milliseconds, and the test asserts it.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Coordinator ruling (I6): until ServiceLevel consumes IOpcUaEndpointState (F-C1), the probe must not sit on
the Active tier of a fused node, because Traefik routes the AdminUI on /health/active. Driver-only nodes keep
it on Ready + Active; admin-only nodes do not register it. Pinned per role.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
ServerAuthenticationStateProvider throws InvalidOperationException from GetAuthenticationStateAsync in any
scope where no circuit set its state (minimal-API requests, hosted-service scopes). The accessor now treats
that as "no circuit identity" and falls back to the HTTP principal. Pinned with the real provider on both
sides (unset -> HTTP fallback; set -> circuit principal wins).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The hand-rolled bearer handler is replaced by AddJwtBearer (Microsoft.AspNetCore.Authentication.JwtBearer
10.0.7, pinned on the same servicing line as the other AspNetCore packages). Validation parameters come
from JwtTokenService.BuildValidationParameters, which now pins ValidAlgorithms = HS256 for both the
bearer scheme and TryValidate; MapInboundClaims is off so the minted Role/zb:username keys resolve.
/auth/token requires the cookie scheme, so a bearer can no longer mint a fresh token for itself.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
DependencyMuxActor retained the last value of every ref ever registered,
and never removed the entries. The continuous-historization recorder
registers every historized ref, so on a historizing node the map grew
with the historized address space and never shrank. The class remark
claimed a bound that did not hold.
- Retain only refs that at least one subscriber registered with
ReplayRetained: true names (VT actors, the Calculation mux adapter, the
scripted-alarm host). A non-opting subscriber is still delivered to but
retains nothing.
- When a ref's last replay-opting subscriber leaves, its value is kept
for a grace (default 2 min, Props(retentionGrace)) so a restart
(old adapter gone, new one not yet registered) still replays. A
periodic sweep then drops it along with the entry. Entries with no
subscriber and nothing retained are removed.
- New SourceRemoved(DriverInstanceId) evicts the values a removed driver
produced, so its last Good is never replayed. The mux records the
publishing driver per retained ref. No production sender exists yet:
DriverHostActor is outside this lane (follow-up).
- Remark rewritten. Tests: retention bounded to replay-opting refs,
release after grace (with an in-grace replay as the positive control),
SourceRemoved eviction, and the restart-replay scenario still green.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The writer concurrency fake parks on its release signal (the 300 ms delay could release the
gate early on a starved pool), the supervisor test's cross-thread hold flag is Volatile, and
the EffectiveTimeout doc now states the 1 s floor the code and tests implement.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Review fix round 1 for lane F:
- The ScanState probe watcher is seeded at InitializeAsync from a platform-only
DiscoverHierarchy query (IGalaxyPlatformSource / GatewayGalaxyPlatformSource over the owned
repository client) and re-seeded on every deploy-watcher rediscovery. The runtime never
calls DiscoverAsync on a live driver, so per-platform host status was unreachable in
production. A failed fetch logs a Warning and never fails init; re-seeds coalesce so a
deploy during an in-flight pass still gets a pass that starts after it.
- EnsureEventPumpStarted refuses while the runtime is stopping, BEFORE reusing the live pump
(that pump belongs to the dying subscriber), and the production rebuild disposes any stale
pump before wiring the new subscriber.
- TeardownRuntimeAsync closes the MX session under the caller's token linked with a 5 s
SessionCloseDeadline, so a blackholed gateway cannot hold DriverInstanceActor.PostStop for 30 s.
- A runtime generation counter, bumped by every teardown, is re-checked before each Rebind
in ResubscribeAllAsync, so a replay-gap re-snapshot that straddles a teardown cannot
resurrect handles into the cleared registry.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
The SDK's DisposeAsync closes with no token, i.e. under the 30 s DefaultCallTimeout, so a
blackholed gateway parked the driver's teardown for the whole of it. CloseAsync(ct) runs the
close under the caller's token, drops the channel on a cancelled/failed close (the gateway
reaps the session on its idle lease) and leaves the session re-openable. A close followed by
the final dispose closes the gateway session once.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- I1 ScriptRunGuard: a run refused because the previous one is still
executing is a Bad publish plus otopcua.script.run_refused, and no longer
a quarantine strike. Only a timeout (and an allocation breach) counts;
a refusal neither adds to nor resets the timeout streak; a completed run
(value or throw) resets it. On a 1 Hz trigger with a 2 s budget a single
4 s overrun is one timeout plus two refusals and must not quarantine.
Tests updated across the guard, the alarm engine, the Calculation
evaluator and the VT evaluator, plus
Refusals_neither_count_toward_nor_reset_the_timeout_streak.
- I3 ScriptedAlarmEngine: the quality bucket is tracked only after the
predicate completes (or in the inputs-not-ready path), so the tracked
bucket is the last one annotated. A refused or quarantined predicate
used to re-track the input bucket Good before the attempt and then flip
it Bad, emitting QualityChanged on every push. The run-guard tests now
count emissions (exactly 1 across N pushes, refused and quarantined
cases), with a healthy sibling's dispatch count as the positive control
that every push was processed. Both were red before the fix.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- RunWriteAsync wraps the invoker call in WaitAsync(writeTimeout) (Props,
default 10 s = 2x the 5 s write cap): a token-ignoring hung write now
answers BadTimeout, releases its slot and tag, and its orphan outcome
is observed and discarded - later writes to that tag reach the device.
- WriteCompleted carries the launching incarnation (from the process-wide
generation source); a late completion from a previous incarnation after
a supervision restart is ignored instead of freeing the new
incarnation's slot/tag.
- Test: a hung token-ignoring tag-a write is abandoned and a later tag-a
write reaches the device while the abandoned one is still inside it.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- GatewayHistorianDataSource.ReadRawWindowAsync keeps the cap + 1 probe budget
(budgetCap) separate from the client's cap: a strict-ceiling or crowding
refusal now reports Limit = the server cap that applied and advises setting
NumValuesPerNode at or below it (it said "lower NumValuesPerNode" with
Limit = cap + 1 to a client that sent 0).
- Fuzz: the oracle-source over-cap assertion was vacuous (cap 0 never counts);
it now checks no reverse window reached the forward-only shape. Class docs
name the post-floor cap set (3 ... 1000) and distinguish NumValuesPerNode
1 / 2 from the smallest server cap.
- HistoryPaging.IsFullPage doc says it judges the REQUEST cap (page + ties +
lookahead), which is finite under any limit; AdvertiseMaxReturnDataValues
swallows any exception (returns false for the caller to log), as its doc
promised.
Refs #736
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
BoundedApiKeyStore's negative cache was keyed by key id, so a flood of DISTINCT well-formed
otua_<random>_x credentials missed it every time and cost one SQL query per attempt under the
SDK's SessionManager-wide lock, no credential needed (R03:3.3).
- IApiKeyStore.ListKeyIdsAsync (default null = unsupported; ConfigDbApiKeyStore lists every
ApiKeyId): the bounded store keeps a known-id set refreshed every StoreCacheSeconds
(single-flight, failed refresh keeps the previous set) and refuses an unknown id without a store
read ({outcome=unknown_id}).
- A process-wide miss budget (token bucket, Security:ApiKeys:MissBurst=100 /
MissRefillPerSecond=20, ApiKeyStoreGuardOptions) bounds lookups that reach the store and miss
while no set is available; a hit refunds its token. Per-endpoint accounting needs the
OpcUaServer impersonation seam to carry the client address — follow-up.
- docs/security.md: the new knobs, and the known-cluster gate's documented fail-open window at
central boot (R03:4.2) plus the async PSK refresh (R03:3.4).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- Writes: a second write to a tag already inside the driver waits in the
queue behind it (_tagsInFlight) while other tags still run under the
cap of 4; queued writes are stamped and failed BadTimeout at dequeue
once older than the 8 s host Ask budget (Props queuedWriteBudget)
instead of being written late.
- Initialise attempts are bounded pessimistically (initializeTimeout,
default 30 s = 2x Tier C) on top of the invoker's token-only timeout:
an overrun is logged as abandoned, reported InitializeFailed so the
retry loop continues, and its late result is dropped by generation.
Generations come from a process-wide Interlocked counter so a prior
incarnation's late result cannot collide (F6).
- PostStop shares one shutdownTimeout budget across shutdown + dispose.
- TelemetryDialSupervisor cancels the establishment deadline on
StreamStopped; FleetDiagnosticsClient builds the ActorSelection inside
the try (a malformed node degrades to "unavailable"); the diagnostics
page labels the column "Last health publish".
- Tests: same-tag serialisation with a distinct-tag positive control;
stale queued write BadTimeout; token-ignoring hung init -> later
attempt connects; stopwatch assertion demoted to a ceiling.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- MeshGrpcCommandDialer.RetainOnly (IMeshCommandDialer, default no-op): CentralCommunicationActor
hands the dialer the current endpoint set on every contact refresh, so a renamed host or moved
port releases its cached GrpcChannel and idle HTTP/2 connection instead of living for the
process (R03:1.3).
- IMeshPskProvider.RefreshAsync (default = the old InvalidateAll): MeshPskProvider pre-resolves
every known cluster's key concurrently, bounded like the sync path, and swaps an immutable
snapshot in atomically; a failed read keeps that cluster's previous key. The actor calls it from
ContactsLoaded (after SetKnownClusters, off-actor via PipeTo) instead of InvalidateAll, so the
first ack per cluster per ContactRefreshSeconds no longer blocks a gRPC thread on a 3 s-bounded
store read or gets denied past the bound (R03:3.4).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
AuditWriterActor documented per-row dedup on the UX_ConfigAuditLog_EventId index but flushed one
statement batch: a single duplicate EventId (any at-least-once producer across a singleton
hand-over) threw for all 500 rows (R03:2.6). SaveChanges now detaches the entries a
DbUpdateException names (2601/2627 logged at Debug as the expected duplicate, anything else at
Warning) and re-saves, bounded by the batch size; a failure naming no entry stays a whole-batch
failure.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Every deploy stored the whole gzip'd fleet snapshot in Deployment.ArtifactBlob forever (R03:1.2).
DeploymentRetentionActor (admin singleton, proxy-less) sweeps every Deployments:Retention:
IntervalMinutes (60): terminal rows (Sealed / PartiallyFailed / TimedOut) older than the newest
KeepSealed (50) and not referenced by a NodeDeploymentState started/applied within ReferencedDays
(30) get their blob EMPTIED — never deleted, never NULLed: the column stays NOT NULL and every
reader already treats zero bytes as "no artifact" (#485), so no schema migration and no reader
change. In-flight rows are never touched; MaxRowsPerSweep (200) bounds one sweep.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- R02:H1 ScriptedAlarmHostActor gains a factory Props
(Func<(upstream, engine)>) so every incarnation builds and owns its own
engine; a supervision restart no longer re-creates the actor around a
disposed engine. The legacy instance-taking Props stays source-compatible
and is now restart-safe in the weaker sense (PreRestart marks the stop as
a restart, PostStop then detaches instead of disposing). DriverHostActor
still spawns through the legacy overload (not owned by this lane — F3).
- R02:H3 OnAlarmCommand runs every engine op under a 10 s token linked to
the actor's lifetime (DefaultCommandTimeout / Props commandTimeout) and
logs a Warning on timeout instead of suspending the mailbox on a stalled
store.
- R02:L3 DependencyMuxTagUpstreamSource.RetainOnly(loaded dep-ref union),
called after every successful load.
- R02:P6 VirtualTagActor's per-failure ScriptLogEntry is rate-limited per
tag (ScriptLogRateLimiter, 20/min + one summary); the Bad publish and the
metric are untouched.
- R07:H6 DependencyMuxActor retains the last (value, status, ts) per
ever-registered ref (through unregisters) and replays them in one batch to
a RegisterInterest(ReplayRetained: true) — VirtualTagActor, the Calculation
mux adapter and the alarm host opt in; the historization recorder does not
(a replay there would record a stale sample per deploy). CalculationDriver
records inputs regardless of _running so a replay landing before
InitializeAsync still satisfies the change-gate.
- R07:P4 CalculationEvaluator negative-caches compile failures by source
hash (one Roslyn attempt + one Warning per source until redeploy); the
driver's script-log entry fires on the Good→Bad transition or a change of
reason, matching the once-per-transition Bad publish.
- R07:H8 array results dedup structurally.
- R07:L2 ShutdownAsync drops the compiled scripts (collectible ALCs).
- R07:U5 DraftValidator rule CalculationNoEffectiveTrigger: a tag with no
timer whose change trigger is off or whose script has no literal
ctx.GetTag read can never evaluate.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- ApiKeyOptionsValidator (ValidateOnStart on any driver/admin node): a blank or placeholder
Security:ApiKeys:Pepper fails host start instead of rejecting every key after a green boot (R03:2.3).
- JWT is finished and opt-in by key (R03:2.4): with Security:Jwt:SigningKey set, a Bearer scheme
(JwtBearerAuthenticationHandler over JwtTokenService.TryValidate / BuildValidationParameters, fail
closed) accepts the tokens /auth/token mints; the default scheme is a cookie-or-bearer policy scheme
and the fallback policy names it, so /api/* and every RequireAuthorization endpoint take a browser
session OR a machine token. JwtOptions are validated at start (JwtOptionsValidator). Without a key,
/auth/token is not mapped (404, not the former constructor 500) and no bearer scheme exists.
A direct AddJwtBearer needs Microsoft.AspNetCore.Authentication.JwtBearer in
Directory.Packages.props (not this lane's file) — the handler is a drop-in to swap.
- AuthenticationStateAuditActorAccessor is the registered IAuditActorAccessor: the Blazor circuit's
principal first, HttpContext as the fallback (R03:2.5). HttpAuditActorAccessor shares its resolver.
- CookieAuthenticationStateProvider (unregistered, fire-and-forget ping loop) is deleted (R03:1.5/2.5).
- /auth/login logs the LDAP exception at Warning before answering 503 (R03:2.7).
- Admin operations reply with a fixed message + correlation id instead of ex.Message (R03:4.1).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- R05:1/12 ScriptedAlarmEngine: _valueCache is cleared in LoadAsync (it is
re-seeded for the referenced paths) so retired tags no longer accrete across
deploys; OnUpstreamChange no longer writes the cache after Dispose.
- R05:3 ScriptedAlarmEngine.Dispose drains in-flight work for at most 5 s
(DisposeDrainTimeout), then cancels it through a shutdown token every
DrainDirtyAsync/ShelvingCheckAsync now runs on, logs a Warning and gives
one second of grace — a stalled IAlarmStateStore.SaveAsync can no longer
hang host shutdown. A save that completes inside the bound is still drained
(positive control test).
- R05:6 ScriptedAlarmSource logs a throwing OnAlarmEvent subscriber at
Warning (naming the alarm) instead of swallowing it.
- R05:7 LocalDbStoreAndForwardSink keeps a cached dead-letter counter in step
with every dead-letter/revive/purge path and resyncs it with the queued
counter on the same cadence; GetStatus no longer runs a COUNT(*) scan.
- R05:9 ScriptLogRateLimiter (20 per source per minute, fixed window, one
summary entry naming the dropped count when the window rolls) behind
ScriptLogTopicSink (key = script|vtag|alarm) and ScriptLogCompanionSink
(key = script name).
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
- NodeManagerHistoryReadRawCapFuzzTests: 10k seeds (env OTOPCUA_PAGING_FUZZ_ITERATIONS)
through a real SDK node manager against wire-faithful fakes of SqlServer / QuestDB /
Gateway that enforce the backend half of the cap; random series with ties, random
windows (empty, on/past a tie, between samples, forward/reverse/end-only, bounds),
N in {0,1,2,small,huge}, cap in {off,3..1000}, tie budget {3,8,65536}. Pages must
concatenate to the cap-off one-shot answer (checked against the series), terminate
within values + 2 (+ one empty step per oversized boundary cluster), never exceed
the cap, never ask the backend for 0 / over-cap; refusals only past the budget.
- HistoryPaging.AdvertiseMaxReturnDataValues writes HistoryServerCapabilities/
MaxReturnDataValues (i=11273), read back through the SDK master node manager.
- MinMaxRawValuesPerRead raised to 3: at 2 a resumed page has no room (value + its
boundary tie + lookahead) and every value cost an empty page + CP (seed 15).
- RawPageLimit rows pin the default-cap arithmetic (9 999 first page, 9 998 resumed).
Fuzz findings recorded as follow-ups: a bounded read whose data fits one page
returns cap + 1 (end bound fitted against N, not the cap; node manager, F2);
the Gateway walk-back's cost on a sparse series under a tiny tie budget is
pre-existing (seed 320: 82 reads uncapped, 93 capped).
Refs #736
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
A failed SDK start (port in use, PKI unwritable, bad OpcUa:* override) was one Error line: the
node stayed /health/ready 200, joined its mesh, could be elected driver Primary and acked deploys
through the Null sink while every client got connection-refused.
- IOpcUaEndpointState / OpcUaEndpointState: the start outcome is a fact the hosted service
maintains (NotStarted / Started / Failed / Stopped + reason).
- OpcUaEndpointHealthCheck ("opcua-endpoint", Ready + Active) is Unhealthy unless Started;
registered by AddOtOpcUaHealth(hasAdmin, hasDriver) on driver-role nodes.
- A driver-only node rethrows the start failure (Program.cs's fatal handler exits non-zero so
the supervisor restarts it); a fused node keeps its AdminUI up but is visibly not ready.
- StopAsync detaches the sinks and then STOPS the SDK (the port closes, clients get
connection-refused rather than a stale, inert endpoint); DisposeAsync is idempotent with it.
- The ServiceLevel calculator / redundancy election live in Cluster + Runtime (not this lane);
the fact is exposed for the Wave 3 follow-up to consume.
Tests: port-in-use ⇒ state Failed, probe Unhealthy, real /health/ready and /health/active 503
(positive control: 200 once Started); driver-only rethrows; StopAsync closes the port and
DisposeAsync twice is a no-op; the probe registration is on both tiers for driver nodes only.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
DriverInstanceDiagnostics.Name carries the string driver-instance id
(what /hosts, telemetry and the artifact key on) and DriverInstanceId
the DB row Guid; the wire record has no slot for the authored display
name. The write-cap test asserts entry order per slot pair instead of
completion order of two concurrent writes.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
R05:10 — the namespace deny-list covered System.Threading.Tasks but not
System.Threading, so SemaphoreSlim / Monitor / SpinWait / Mutex /
EventWaitHandle could park a script's pool thread past its budget (or touch a
system-global kernel object); System.Type.GetType(string) / GetMethod* /
InvokeMember and System.Runtime.CompilerServices.RuntimeHelpers resolved and
reach a forbidden type without naming it syntactically.
ForbiddenTypeAnalyzer now denies the whole System.Threading namespace with a
type-granular allow-list (CancellationToken, Interlocked, Volatile — the
sync-free primitives), adds a member deny-list for System.Type's reflection
entry points (GetType/GetMethod(s)/InvokeMember/GetProperty(ies)/GetField(s)/
GetMember(s)/GetConstructor(s)/GetEvent(s)/GetDefaultMembers; instance
object.GetType() stays allowed) and puts RuntimeHelpers on the type deny-list.
Thread/ThreadPool/Timer leave the type list — the prefix rule reaches them.
R05:11 — the allocation budget landed with E1's ScriptRunGuard
(Scripting:MaxAllocBytesPerRun, default 64 MB): sampled around each run,
breach quarantines. It is observation after the fact; a pre-emptive OOM guard
still needs an out-of-process runner (follow-up issue in the lane report).
The Monaco editor shares the analyzer, so DiagnoseTests pins every new form
as an OTSCRIPT_FORBIDDEN marker and every allow-listed form as clean.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
DriverHostActor wraps the health publisher it hands its children in a
ChildHealthLedger (last snapshot per driver id, forwarded to the real
publisher), so GetDiagnostics now answers with each child's row id,
authored name, last-published state, device counts (host statuses when
the driver probes hosts, else one device) and last-change time — instead
of Guid.Empty / "Spawned" / 0 / now. Entries are forgotten on StopChild
and operator Restart.
IFleetDiagnosticsClient gains its first consumer: the new
/fleet/diagnostics page (node selector + ?node= query, revision,
running-from-cache notice, per-driver table). FleetDiagnosticsClient
logs the degrade-to-empty path at Warning.
Review ref: 02/U1.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Review 2026-10-21 G-5, G-6, G-8 ("wire them all").
- G-5: GatewayGalaxyHierarchySource now pages the DiscoverHierarchy RPC
itself (DiscoverHierarchyRawAsync) at Repository.DiscoverPageSize. The
mxaccessgw client's typed DiscoverHierarchyAsync overloads hard-code a
5000-row page and DiscoverHierarchyOptions carries no page size, so the
authored knob reached nothing; the raw request takes page_size on the
wire, so it is honoured here (non-positive → 5000). The loop keeps the
client's repeated-page-token guard.
- G-6: GalaxyDriverProbe reads ProbeTimeoutSeconds from the config and
runs under min(caller budget, configured cap) — the knob can shorten a
probe, never extend it past the host's budget. The TCP preflight now
shares that deadline instead of hanging on the caller's token alone.
- G-8: GatewayGalaxyDataWriter runs a batch as one lane per distinct tag,
concurrently and bounded at 8 (MaxConcurrentWriteLanes); same-tag
writes keep their authored order inside one lane and results land at
their request's index. Acks stay serial by design.
- GalaxyDriverForm: help text now says what the two knobs actually do.
Refs #735
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
ReceiveAsync suspended the mailbox for the whole of every handler, so one Test Connect against a
black-holed PLC (60 s clamp) or one large deploy compose stalled every other admin operation
fleet-wide — AdminUI alarm acks/shelves, driver restarts, the next deploy.
- TestDriverConnect runs the probe on the thread pool and pipes a TestDriverConnectResult
straight to the requester (every path yields a result, never Status.Failure).
- StartDeployment serialises only the in-flight check (DB row + an actor-local compose-in-flight
flag, because the Deployment row is written at the END of the compose); the snapshot read,
validation, gzip and row write run off-actor and return as DeployComposed /
DeployComposeFailed, where the coordinator is told and the requester answered.
Tests: an alarm ack completes in < 1 s while a 60 s probe is pending and while a compose is
parked on a gated DbContext (positive controls: both still complete afterwards); a second
StartDeployment during an in-flight compose is refused.
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
A timed-out script is abandoned, not stopped, and every seam re-ran it on the
next trigger, leaking one pool thread per trigger (R03:1.1, R05:2, R07:L1).
A shared ScriptRunGuard, keyed per script instance, now:
- refuses a run while that script's previous orphan is still executing
(Bad "previous evaluation still running" + otopcua.script.run_refused);
- quarantines after 3 consecutive non-completions or an allocation-budget
breach (sampled with GC.GetAllocatedBytesForCurrentThread; default 64 MB):
no further dispatch, Bad quality, one Error, otopcua.script.quarantined,
cleared on source change / redeploy / reset;
- starts the budget clock when the pool picks the run up, so starvation is
no longer charged to the script (R07:P5), and cancels the evaluator token
on timeout.
Wired into RoslynVirtualTagEvaluator, CalculationEvaluator (new async
EvaluateAsync; the driver's pump and timer ticks now await the run behind a
SemaphoreSlim gate) and ScriptedAlarmEngine (a non-completing predicate
holds state and annotates the condition BadInternalError). ScriptingOptions
(Scripting:MaxAllocBytesPerRun, Scripting:QuarantineAfterConsecutiveTimeouts)
carries the knobs; the Calculation driver config accepts the same two keys.
Tests: guard unit suite (incl. the 30-trigger runaway leaving one orphan with
a healthy sibling as positive control), per-seam refusal/quarantine/reset
tests. The S12 race test now uses one vtag id per loop: twelve concurrent
evaluations of ONE id is a shape production never produces and the guard
refuses by design.
Refs #734
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
CallTimeoutSeconds was bound, defaulted and documented but consumed
nowhere. TelemetryDialSupervisor now arms a per-generation deadline on
every dial: a dial that has delivered no event by the deadline is
cancelled and treated as a transport failure (reconnect with a failure
streak). A live node replays its snapshots on subscribe, so this bites
only a dial that never connects (a black-holed host, where the socket
connect has no timeout). An established stream is deliberately not
bounded — the wire has no heartbeat, keepalive pings own its liveness.
Review ref: 02/U2.
Refs #731
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
StartDeployment read the whole config DB twice with no transaction — DraftSnapshotFactory for
the validator, then ConfigComposer for the artifact, plus clusters/nodes and Script.SourceCode
again in between. An AdminUI save landing between the two shipped an artifact the DraftValidator
never saw.
- DraftSnapshot carries ServerClusters, ClusterNodes and ApiKeys; DraftSnapshotFactory reads
every table inside one SNAPSHOT transaction on a relational provider (no-op on InMemory or
under an ambient transaction).
- ConfigComposer.Compose(DraftSnapshot) is pure over the snapshot (ordinal-sorted keys);
SnapshotAndFlattenAsync(db) delegates to it, byte-identical.
- AdminOperationsActor validates topology, estimates compile cost and composes the artifact
from the one snapshot it holds.
- Migration EnableSnapshotIsolation: ALTER DATABASE CURRENT SET ALLOW_SNAPSHOT_ISOLATION ON
(suppressTransaction; no model change).
Tests: the SQL Server snapshot read ignores a row committed mid-read (positive control: a later
read sees it); an edit injected between the Equipment and Tags reads is absent from the shipped
artifact and rejected by the next deploy; Compose(snapshot) == SnapshotAndFlattenAsync(db).
Refs #733
Claude-Session: https://claude.ai/code/session_01Ub9hGmYLQvn9XETzw7CMsT
Joseph Doherty