IME's trace listener opens every log once per process with FileMode.Append and FileShare.ReadWrite
and writes at its own stream position afterwards. AgentExecutor.exe (two processes per user-context
script, one per detection script) has such a listener on AgentExecutor.log and on
IntuneManagementExtension.log, so a process that resumes logging after another one appended
overwrites those bytes: the final file holds the last writer, a byte-position tailer has read the
first version. Session 46749560 lost 6 of 16 platform-script results and 4 of 16 executor end
blocks with every tracker health counter at zero — results fell back to the exit code (one Success
where IME said Failed), one finished script stayed Incomplete. The agentexecutor_fallback results
in the fleet are this race, not late IME logging.
The tracker now keeps a fingerprint ledger of every entry it processed from the two multi-writer
files (FNV-1a 64 over the decoded bytes, one CmTraceEntryAssembler shared by the read loop and the
verification so entry boundaries are identical by construction), verifies it on every script end
signal, on a byte-budgeted cadence while an invocation is in flight (256 KB/s, at least 1 s
apart), in a 4 KB guard zone on every growth pass and on a fragment at the bookmark, and rewinds
to the first changed entry; entries whose (offset, hash) still read unchanged are skipped,
everything else is processed as backlog (never fresh, never a calibration anchor). The
policyId-less executor lines are owned by the nearest preceding invocation marker in their file
(PS-AGENT-ARG opens, PS-AGENT-SCRIPT-START upgrades to the platform script, PS-AGENT-INVOCATION and
PS-AGENT-COMPLETED close) instead of a last-started pointer, so an end block a rewind uncovers
reaches its own script. Restart seeds the ledger from the 64 KB behind the persisted bookmark;
a rollover clears it. Four restart-safe counters (ime_overwrite_rewinds,
ime_overwrite_bytes_reprocessed, ime_verify_passes, ime_verified_bytes) ride agent_metrics_snapshot
and ime_pattern_hits — recovered data, not skipped work, no new event type.
Tests: ImeLogTrackerOverwriteRewindTests (15) drive the positional-overwrite write pattern against
both files, including the two field scenarios with the shipped patterns; agent suite 2921 green,
backend known-actions list extended for the two new patterns.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
SecurityValidator stamps ValidatedTenantId once the certificate tenant binding
(or the bootstrap token) proves the tenant, and DeviceValidation once the
device-validation chain has run: the admitting ValidatorType name, None for a
tenant without a validator, Transient for the 503, Rejected for the 403 of
stage device. RequestTelemetryMiddleware resolves the row's TenantId in one
place (policy tenant, proven tenant, then the X-Tenant-Id header only on the
routes that validate it) and copies DeviceValidation as a dimension.
Register, config and upload-url rows carried no TenantId before, so
tenant-filtered KQL missed exactly the calls that run the validation, and the
validators' success line is Information-level and never reaches App Insights.
Proving which validator admitted a device is now one query.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- GetMcpOrganizationUsageResponse.daily: charged requests per day inside the range from the same McpTenantUsage rows (BuildOrganizationDaily, MSP reads included); manifest + wire types regenerated (web + MCP copy); fold test
- Web: viewing another tenant no longer fetches the caller's own usage — no UPN, plan badge or own quota; the summary tiles and the daily chart come from the tenant's daily series. The own tenant stays the caller's usage page. Internal gitlink.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- Backend: GET global/metrics/mcp-usage/organization?tenantId= (GlobalReadOrAdmin, TenantScoping.None on purpose — the route enumerates accounts, so the delegated path stays out; 400 without a GUID, no aggregate); GetMcpOrganizationUsageResponse carries `quota` (tenant plan + daily/monthly windows summed from the same McpTenantUsage rows); McpQuotaService.ResolveTenantPlanAsync with public TenantPlanLimits; two folds tests
- Web: TenantScopeSelector on /settings/reporting/mcp-usage (override-only, hidden for a delegated scope), scopedApi.mcpOrganizationUsage, Organization-quota card from the organization response (falls back to the caller's own node only on the own tenant), latest-wins fetch guard, wording follows the viewpoint
- Manifest + wire types regenerated (web + MCP copy); internal + skills gitlinks
Deploy order: Backend before Web — under a tenant override the page calls the new route.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
An app still installing when the session's observation ended has no terminal
event and never will. The session page ticked Date.now() - startedAt for it
("11h 39m" on a session that finished hours ago), and its AppInstallSummaries
row stayed InProgress forever - ranked behind failures in the app's session
list and counted in totalInstalls with no named bucket.
Web: applyObservationEnd marks such rows Incomplete once the session is
terminal (isTerminalStatus), bounded by session.lastEventAt; the panel shows a
slate Incomplete badge and the watched span as a lower bound, the "active"
pill excludes them, and the download panel carries the parallel flag. Covers
every historic session without a backfill.
Backend: CloseOpenAppInstallsForSessionAsync closes InProgress rows without a
CompletedAt as Incomplete - at the terminal seam, again on the ingest batch
that carries agent_shutting_down for a terminal session (rows opened after
the verdict), and from the 30-day sweep. The sticky-status reconcile keeps
Incomplete against a late progress-only batch and lets a real terminal win.
App metrics and app analytics disclose the bucket (incomplete,
totalIncomplete) outside the failure rate; manifest and wire types
regenerated.
MCP: get_session_summary coverage gains an apps block and a gaps line naming
the apps the agent's last app_tracking_summary still listed as installing or
downloading, terminal sessions only.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Closed submissions leave through the daily maintenance sweep (RuleSubmissionRetention, one place for rule and test); pending/approved/published never. The submit dialog takes an optional contact email (a UPN is not always a mailbox), stored for the reviewer and never published. The tenant list shows pending rows by default with the rest behind Show all, and the contribute button shows for every writer, also before the first custom rule exists. DELETE global/rule-submissions/{id} (GlobalAdminOnly, MCP delete_rule_submission) removes test/demo rows for operator and tenant alike. Bumps internal and .claude/commands gitlinks.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Tenant admins submit references to their own gather/analyze rules; the server freezes the rule, records pre-flight findings and fire stats, and stores it under a 12-character id (RuleSubmissions, SessionReports layout). Attribution is the submitter's frozen choice (anonymous by default) and becomes the published author. Review (approve with reserved id / decline with comment) through /admin/reports/rule-submissions or the MCP tools list_rule_submissions, get_rule_submission, review_rule_submission, reseed_rules_from_github; published is derived from the global catalog, never stored. Analyze rules now get the same server-side author stamp as gather rules; ANALYZE-ID-001 carries author Community. Privacy page names the credit; bumps internal and .claude/commands gitlinks.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
ANALYZE-ESP-004 fired on every ESP terminal failure of a Continue-anyway
profile and rendered the configured FirstSync limit as elapsed time — the
mayHaveContinuedAnyway flag only says the profile has the button. The rule
now requires the agent's own verdict (app_install_failed with
failureType=esp_apps_timeout), names that app via a same-event filter, reads
failedSubcategory from esp_failure_settle_started and words the limit as a
limit. Verified with test_analyze_rule: cb7036a6 fires, 683f1eff (0x80070652
after 19 min) stays silent.
Rule engine: not_exists with a dataField on event_type/event_data
conditions matches when no event of the type carries a non-empty value —
including an absent type — instead of being dead behind the fieldValue null
short-circuit. Schema and MCP authoring guide say so.
Agent: a Shell-Core 62407 failure that lands inside the provisioning
tracker's 30-s settle window (a third of all ESP terminal failures) now
inherits the pending errorCode / failedSubcategory / category /
likelyCulpritApps, so enrollment_failed, Sessions.FailureReason and the
termination-time app classification see the HRESULT the registry had
already named. failureType stays the Shell-Core value.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The What's new panel prints "Latest agent version" under the release
entries, and the session list draws its outdated badges from the same
endpoint — but the value was cached 12h in each Function instance and 1h
in the browser, so the panel announced the previous release for up to 13h
and two instances could disagree.
- LatestVersionsService: TTL 12h -> 5 min; keep the last successful result
and serve it when the blob read fails, so the shorter TTL cannot blank
the version line on a transient hiccup (failure marker 5 min -> 1 min).
- GetLatestVersions: Cache-Control max-age 3600 -> 60.
- No refresh callback from build-agent.yml: the cache is per instance, a
warm-up request would leave every other instance stale.
The origin is a ~200-byte no-cache blob behind the download alias that
Front Door does not cache, so a warm instance now reads it at most 12x/h.
Backend deploy (workflow_dispatch) pending — B-uq7.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
scripts/ at the repo root is the customer surface: everything under
Bootstrap/ and CustomerSetup/ is published to the download container by
Publish-BootstrapScripts.ps1, Authenticode-signed and version-gated. The
error-code generator is a maintainer tool no workflow runs, so it diluted
that invariant. .github/scripts/ was not the alternative -- it holds only
helpers a workflow step invokes, and unlike guardrails-in-sync and
shared-manifests-in-sync the catalog has no CI drift gate (the generator
needs IME_ENUMS_JSON from operator tooling and live MS Learn fetches).
REPO_ROOT stays ../.. at the new depth, so only the documented path moved:
the generator header and README, the C# and TypeScript comments, the test
summary, and the description header of error-codes.json together with the
generator literal that emits it (the two must stay identical or the next
regeneration diffs). The web sync re-ran clean at 607 entries with no
change to utils/error-codes.json, which drops the header.
The card composed its line from the backend's explanation ("did not answer within
3s - it scales to zero when idle ... ~20-30s"): probe mechanics nobody waiting on a
starting instance needs. It now writes its own one-liner and ignores check.message,
so the server's "warming" and a client-side probe timeout read identically and the
wording does not depend on a backend deploy. The backend message is shortened but
still carries the budget for direct API callers.
The re-poll budget of 8 x 5s (~40s) expired on a real cold start - Container Apps
activation has no upper bound we control - so it is now 12 x 5s (~60s), with a test
pinning that the stated seconds come from cadence x attempts.
Measured over 14 days, GET health/mcp answered in 4..148ms when warm and timed out at 30s
in 50 of 83 calls — the median call took 27,972ms. A cold start is 20-30s of Container Apps
activation (~98% KEDA plus image mount, ~500ms application), so no UI wait covers it and the
generous timeout only bought a long wall in front of the same answer. The browser's own
budget is the same 30s, so the backend's carefully worded warning never reached the user at
all: they saw a raw DOMException.
Backend: the budget drops to 3s, which sits in the widest observed latency gap
(2347ms..3930ms) so jitter cannot flip the verdict. Our own cancellation now reports the new
status "warming" at Information level (with minReplicas=0 a cold probe is the expected path,
and logging it as a warning poisons the ops signal); HttpRequestException is split out of
that catch and stays a real warning, still without the host in the message. The warm-hit
threshold drops to 750ms — ~5x the warm maximum, deliberately generous because every
measurement comes from a near-idle single-user system and warm latency will rise under
concurrency — and moves into McpReachableMessage so a test can pin the boundary. The XML doc
now describes the probe that exists rather than the wake-signal design the measurements
disproved.
"warming" is a new status value rather than an added DTO boolean because the backend deploys
before the web: an unknown value degrades to a grey card with a clean banner, whereas a
boolean would leave old web showing yellow and "Warnings detected" on every cold start.
HealthCheck.Status now documents the five-value vocabulary, which rides into both generated
TypeScript mirrors as JSDoc.
Web: resolveMcpCardState owns display, banner rating and the poll decision as a pure
function (the suite runs on node, so testable logic cannot live in JSX). A warming server
re-probes every 5s for up to 8 attempts and stays out of the overall banner exactly as
"unknown" already does — scaling from zero is expected behaviour, not a degraded platform —
and only becomes a real warning once the re-polls are spent. The card keeps showing
"warming" while the next probe is in flight so it cannot flicker, carries the blue
in-progress look instead of yellow, and was hoisted out of the healthResult guard: a failing
detailed check would otherwise hide a card that is still polling.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
The portal hides Delete for owned (msp-{home}) groups, but DELETE global/tenant-groups/{groupId} accepts any id: a Global Admin deleting one ended Pro (MSP) for every member with no grace anchor, so a member that raised its retention under Pro was clamped to the Community cap at the next sweep. DeleteGroupCoreAsync now snapshots the owner and calls ProConferralService.RecordLossAsync(member, owner, "group-deleted") per member, owner excluded, independent of assignees. Two tests in RevokeEnforcementTests; internal docs list the fourth writer.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Act equals the KEV set by construction (CvePriority: KEV => act), so the KEV tile becomes the Act tile and Attend joins it instead of a duplicate count. The breakdown is counted in the backend over every CVE in the window, like the severity breakdown, because the list is top-N. Both panel consumers read the generated CveExposureSummary type; the MCP tool description names the new block. Wire types regenerated for web and MCP.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The caller-supplied X-MCP-Tool-Name header went verbatim into the UserUsageLog
row key ({yyyyMMdd}_{tool}:{endpoint}) and from there into the CAS-conflict
warning of TableCasRetry. EndpointNormalizer.ToolNameKey now strips CR/LF and
every other control character or line separator, replaces the characters Table
Storage rejects in keys (/ \ # ?) and caps the length at 64; an empty result
skips the prefix. Real tool names pass through unchanged, so existing usage
rows keep their keys. A header with an illegal key character previously made
every usage increment fail with 400 (fail-soft, counter never advanced).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The report still carried two hand copies after D-215: a typed map of C#
initializer defaults for the "custom" highlight, and a TypeScript mirror of
GetAgentConfigFunction's resolution rules for the runtime column.
Backend: the derivation moves verbatim into Services/AgentConfigResolver
(Build = pure, ResolveAsync loads tenant + admin config and the two rule
catalogs). GetAgentConfigFunction keeps the kill verdict, the migrate logging
and the device fields on top; ParseAgentMajor, ResolveDiagnosticsUploadEnabled,
MergeDiagnosticsLogPaths and ResolveMigrateTarget move with it (call sites and
tests retargeted; the QueryParams ratchet baseline shrinks by the file). New
GlobalAdminOnly route config/{tenantId}/effective-agent-config returns the same
AgentConfigResponse without a device (optional ?agentVersion=, default the V2
line). AgentConfigResolverTests pin the rules that used to live in the web:
upload 10, Hosted destination without SAS, SAS regardless of mode, unrestricted
only with entitlement + gate + toggle, operator knobs from AdminConfiguration,
analyzer defaults, hash line per major, migrate override and rejection.
Manifest: SharedManifestParityTests emits "defaults" next to "fields" for
TenantConfiguration and AdminConfiguration and an "agentConfig" section with
the AgentConfigResponse / CollectorConfiguration / AnalyzerConfiguration class
defaults (fresh-instance values; nested objects null). TenantConfiguration.
NtpServer gets the initializer "time.windows.com" the table read already
applies, so the manifest default and a stored row agree.
Web: the catalog rows lose value() and default; the runtime column renders
the endpoint's response, the highlight compares against the manifest defaults,
informational rows (reasons, timestamps, identities, secrets) never read as
custom. Previously excluded fields are visible (configVersion, catalog counts,
sealing pattern ids, hash oracle, migrate target); the collector class
defaults sit in a collapsed "Collector Internals" section. Fail-soft: while the
backend is not deployed the runtime column shows a notice.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The "Health by Device Model" link searched "{Manufacturer} {Model}" as one substring; the server's
free-text predicate tested it per field, so once the failed sessions were outside the loaded first
page the sweep returned nothing.
- lib/searchQueryTokens.ts: tokenizer shared by the timeline and the session search (terms, quotes,
leading minus, key=value / key:value qualifier when the separator precedes any quote)
- app/dashboard/utils/sessionSearchQuery.ts: 15 qualifiers = exactly the fields both sides search,
matcher over session fields + client-derived extra text, suggestion field lookup, link builder
- useDashboardFilters / SessionTable: filter and suggestions use the grammar; fuzzy only for a
single free term; session_searched gains a `qualifiers` count
- Helpers/SessionSearchQuery.cs: the same grammar for q= on /api/search/sessions (+ global), parsed
once per request; MatchesFreeText is a thin wrapper
- utils/session-search-syntax.cases.json: one case file run by vitest and by
SessionFreeTextSearchTests (parity proof)
- FleetModelHealth/FleetSlowModel/FleetFailingModel: manufacturer + model as stored plus label,
grouped on the pair; useFleetHealth uses the generated wire types; "Unknown" is not a link
- shared manifest + wire types regenerated for the three DTOs
Backend suite 6158/6159 (the one failure is SharedManifestParityTests against uncommitted WIP of a
parallel task), web vitest 1177/1177, build 0 warnings.
Parallel tool calls of one user hit the same (user, day, endpoint) row in the same instant; without backoff the losers stayed synchronised and 3 of 6 increments were lost. Both increments now use the one CAS helper (merge patch, AddEntity on 404, StorageCasConflict marker) with UsageCounterRetries = 8; the tenant counter's attribution columns travel in the merge patch. Tests pin the budget, the patch shape and the create path.
- IME-EXITCODE-CLASS captures the return-code class of the installer exit code ("lpExitCode is defined as <Success|SoftReboot|HardReboot|Retry|Failed>", verified against the IME string literal) as exitCodeClass on AppPackageState; new action captureExitCodeClass, persisted across the agent restart a reboot causes, last value wins across a Retry loop. IME-EXITCODE-MAPPED-FAILED keeps the error path; both match the same line.
- Adapter: exitCodeClass whenever known, exitCode on every terminal state (Installed/Postponed/Skipped/Error) instead of Error only.
- ANALYZE-APP-018 (warning, apps): app_install_completed.exitCodeClass in SoftReboot,HardReboot; +25 HardReboot, +15 observed system_reboot_detected; never fails the session.
- Web: completed-app exit badge amber (failed stays red), "mapped as SoftReboot/HardReboot/Retry" on timeline and Install Progress.
- Tests: pack end to end (phase, current app, class lines, negatives, restart round trip), adapter payloads, linearity samples, backend rule engine + known actions.
- internal gitlink: plan, D-214, backlog.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
MCP: coverage block (window, IME tracker, collectors, upload, diagnostics, gaps) built from the
health events the summary already fetches (session-coverage.ts, data.<key> slices); validate_rule
warns on a required not_exists of an IME-log-derived event without the ime_tracker_degraded
precondition; authoring guide 'absence'; instructions and prompts point at coverage.gaps.
Rules: guardrails.json imeLogDerivedEventTypes (schema + combine). Backend: catalog invariant test —
absence rules carry the coverage gate or a documented exception (ANALYZE-ID-004).
Agent: diagnostics_uploaded/_upload_failed carry packaging counters (truncated, includedFiles,
skippedFiles, skippedByReason, problemsByKind) — counts only, never paths.
Also: shared-manifests.json regenerated (was stale at HEAD; McpAccessPolicy doc string) + codegen.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
TenantConfiguration.McpDisabled + McpDisabledReason (GA-only, same deny-list and PUT guard as Disabled).
Front door: McpUserService denies every caller of a closed home tenant below the platform roles.
Per request: PolicyEnforcementMiddleware denies a closed home or target tenant for X-Client-Source: mcp
(403 McpDisabled) and narrows the delegated fleet bound, never to null. ClientSourceHeader is the one
reader of the header. The MCP access guard surfaces the backend reason verbatim. GA tenant modal,
list badge and config report; wire types regenerated; internal doc, health marker, gitlinks bumped.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every HTTP request body the web, the MCP or the agent sends is now a flat class in
AutopilotMonitor.Shared implementing IApiRequest — the request twin of IApiResponse.
WireTypeManifestBuilder exports the implementers as a third root, so the request DTOs
reach wire-types.generated.ts on both clients; for request types a nullable property
keeps `| null` (a PATCH may send an explicit null to clear an override).
Functions read a body only through RequestBody.ReadAsync<T>/ReadOptionalAsync<T>
(System.Text.Json, ApiJsonOptions.Read = wire options + case-insensitive; empty or
malformed body = 400 envelope, never a 500) whose generic constrains on IApiRequest.
ReadNewtonsoftAsync<T> stays for the Newtonsoft-bound cluster (tenant/admin config via
the patch service's PopulateObject, rule and pattern documents) as a frozen per-file
baseline. The tenant-plan PATCH keeps its presence-sensitive JsonDocument walk with the
keys derived from PatchTenantPlanRequest by nameof (PlanPatchKeyParityTests).
35 nested request classes, 7 JObject/JsonDocument hand-walks and the anonymous
DeserializeAnonymousType bodies moved to Shared; the two identical TicketRequests are
one DownloadTicketRequest; ServerAction is no longer the request (QueueSessionActionRequest,
the stored row is built server-side); create-from-template takes { "variables": … }
instead of a bare map. Fields the endpoints default server-side are nullable now
(CPE mapping, report attachments, feedback rating on dismissal, upn for applications).
TypedRequestGuardTests: no raw deserializer under Functions/ (named exemptions: agent
routes frozen by D-200, Azure Monitor webhook, upstream/stored-column parses), the
Newtonsoft branch only in its baseline, no request type in the Functions assembly, every
DTO flat and in Shared. RequestBodyTests cover both readers, the cap and the envelope.
QueryParamGuardTests baseline shrinks by 14 (the Content-Length TryParse lives in the reader).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The surface comes from the signed appidacr/azpacr claim (0 = public client = portal;
secret/certificate, app-only or missing = integration), never from a header: portal SPA,
MCP server and API share one app registration in production, so the client id cannot tell
them apart and any token holder controls headers. New AdminConfiguration field
PortalUserRateLimitRequestsPerMinute (default 600, Global Settings round-trip, wire types
regenerated); the existing user budget is relabelled MCP & integrations and keeps the tenant
override and the Pro floor; buckets are keyed per surface; ThrottleSurface is a request
dimension next to ClientSource so the classification is auditable (platform-health check 2).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A freshly onboarded tenant spent 45 minutes in 503 because a second instance held a
roleless app-only token minted before the consent; InvalidateTenant never reaches other
instances. GraphAuthFailure is the shared step for all five Graph validators: attempt 1
drops the cached token and reports transient so the existing loop re-mints; attempt 2
keeps each validator's semantics (Autopilot/DeviceAssociation transient with Retry-After
120, CorporateIdentifier/CloudPc/IntuneBinding definitive) and logs a stable marker.
SecurityValidator hands the agent the largest Retry-After requested. Platform-health
check 6 and the marker catalog know both markers; the cloudpc concept doc carries the
contract.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
About 12 MB/day of severityLevel-1 traces without diagnostic value (MSAL verbose, StatusCodeResult,
CORS success, Application started). Application logs were Warning+ already via Function: Warning.
Measured and documented in internal/docs/backend/telemetry-ingest-shaping.md.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- StorageClientOptions: Table/Queue 8 s per attempt × 3 attempts + capped backoffs = 25.8 s < the
agent's 30 s upload timeout (pinned by test); Blob 60 s × 4. Every production client takes its
options from there. No Polly over the SDK.
- StorageErrors.IsTransient(Exception): OperationCanceled / IO / HttpRequest / Aggregate-of-those
/ RequestFailed status 0; the ingest answers 503 + Retry-After instead of 500.
- The ingest still takes no caller cancellation (a batch is a unit). GetAllSessionsPageAsync
carries the request token down to BoundedFanOut (Global Admin list); SingleFlightCache hands its
factory a 30 s budget token and evicts on timeout (tenant ids, app-summary snapshot, session stats).
- ProcessStallMonitor: 1 s tick drift ≥ 1 s ⇒ ProcessStall event (sampling bypass) + metric with
cause gc / threadpool / unknown from GC.GetTotalPauseDuration and thread-pool counters;
GcPauseMs on request rows ≥ 100 ms. Measures the instance-local 5–10 s pauses seen in production.
- Gitlinks: internal (plan, D-203, concept doc, B-9hf, lesson), .claude/commands (marker rows).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Helpers/QueryParams replaces fourteen private days/limit parsers, the three
TryParseUtc copies, eleven identical pageSize blocks and ~50 inline TryParse sites
under Functions/ and Pagination/. Lenient (Int/IntOrNull/UtcInstant): absent, garbage
or below-minimum means the call site's default, above-maximum clamps. Strict
(TryInt/TryPageSize/TryUtcInstant): absent is null, anything else invalid is a 400
with a message that names the parameter. Integers are invariant-culture, instants are
invariant UTC (AssumeUniversal | AdjustToUniversal), whitespace counts as absent.
Defaults and ranges stay arguments at every call site.
Settled differences: the Apps windows clamp above 365 instead of falling back to 30,
days=0 or negative yields the default everywhere, the two global geo endpoints and
the session list cap days at 365, SLA months is capped at SlaMetricsService.MaxMonths
on the HTTP layer, a whitespace date means no filter on every route, the admin batch
sizes reject values above their maximum, and the three culture-dependent
DateTime.TryParse sites parse as UTC.
QueryParamGuardTests ratchets TryParse out of Functions/ and Pagination/ (baseline:
headers, raw cells, the agent version major, the se= of a stored SAS URL);
QueryParamsTests carries the cases the removed copies pinned.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
TelemetryItemKind moves to Shared; TelemetryItem.ToWire() maps onto TelemetryItemDto
and the uploader serialises the DTO, so both ends share one wire class. The spool line
keeps writing Kind by name (StringEnumConverter registered in TelemetryItemSerializer).
IngestTelemetryFunction.PartitionBatch routes by exact kind name and collects every
item it cannot route or parse; a non-empty rejection list answers the batch with 422 +
TelemetryItemsRejectedResponse (poison, rejectedRowKeys, reason) before anything is
persisted, logs a Warning and writes the new TelemetryItemsRejected ops event. The
agent's existing poison protocol drops exactly the named RowKeys and re-uploads the
rest; before this the ingest skipped the items, answered 200 and the agent cleared
its spool.
Contract tests from both sides: TelemetryWireContractTests (agent) freezes the real
batch body under tests/fixtures/telemetry-wire/ and parses the pinned poison body;
TelemetryWireFixtureTests (backend) dispatches every frozen body with zero rejections,
requires an arm per kind and pins the poison body.
Agent-facing responses implement IApiResponse and go through ResponseHelper, request
types carry [WireContract]; plain ingest refusals use the error envelope, the
register-session and upload-url refusals keep their DTO bodies (the agent reads
ErrorCode/Message). Manifests and generated wire types regenerated (also picks up
four types HEAD had left out of the manifest).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
PlatformStats(global,current) took a GET plus a full-row upsert without ETag
from every ingest batch: increments were lost under concurrency and a writer
clobbered fields another writer had just changed. RuleStats put every tenant's
daily rule traffic into one date partition (1,459 rows on 2026-09-04), the
writer issued four serial calls per evaluated rule without ETag, and the
reader was a cross-partition scan whose 10k cap dropped the newest dates.
- D-198: TotalEnrollments / SuccessfulEnrollments / TotalEventsProcessed have
no hot-path writer; the two-hourly recompute is the only source. Only
IssuesDetected still increments, through TableCasRetry (ETag + field-only
merge, AddEntity for a missing row, conflicts as StorageCasConflict). The
tenant counters share the helper.
- D-199: RuleStats PartitionKey "{scope}_{date}", RowKey ruleId.
RecordRuleStatsAsync merges a session's increments per rule, reads the
partition once and writes one transaction (UpdateReplace with ETag / Add),
retrying on 412/409. Readers are PartitionKey ranges per scope; aggregation
and the regression radar fan out over the registered tenants; the cleanup
deletes per scope. RuleStats joins the offboarding composite wipe.
- Legacy rows (PartitionKey = date) are read by an extra query and cleaned by
a date range that checks every row for a bare-date key, until 2026-12-05.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
- DiagnosticsBlobStreamer disposes the download result on every exit path
(it was disposed only on the 413 branch), measures open and transfer
separately and fires DiagnosticsDownloadProxied after the copy with
Completed=true/false.
- UserDelegationKeyCache: one user delegation key per process (24 h, re-mint
under a lock with 2 h margin or when a SAS would outlive it) instead of one
storage round-trip per download or upload SAS.
- KevDataService is resolved as one singleton over the typed-client handler
pipeline; the cache warmer used to hydrate a throwaway transient instance
while the correlation service held a cold one.
- Admin and tenant configuration services single-flight concurrent cache
misses (Lazy<Task> per key); the MemoryCache and the in-place projection
semantics stay untouched.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The dependency filter dropped every successful storage call and the expected
412/409 outcomes, so storage-side latency and ETag contention were invisible
until a call failed outright (seven days of App Insights held zero storage
dependency rows).
- StorageDependencyFilterProcessor keeps a successful storage call at or above
2 s and stamps SlowStorage=true; the operation name is on the row, the table
name is not (the Azure SDK span carries neither URL nor table).
- StorageMetrics wraps TelemetryClient.GetMetric("StorageCasConflict",
operation, table, outcome): pre-aggregated per minute, bypasses the worker's
adaptive sampling, null-tolerant for tests.
- The CAS loops report conflicts: session event count, session counter
reconcile, session store, tenant stats, MCP user/tenant usage.
- TableStorageService keeps its two-argument constructors (Moq matches by
exact arity); DI resolves the three-argument overload.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every entity-group transaction was chunked by count only; 100 payload-heavy rows
(Signals with chunked PayloadJson, Events with 30k-char Message + DataJson) could
exceed the 4 MiB transaction cap, and the failure surfaced as an opaque 500. The
agent halves its batch only on 413 and treats 5xx as "replay later", so such a
batch wedged the device forever.
- TableTransactionBatcher: 100 actions, 3.5 MB per transaction, 1 MB per entity,
never across partitions; an entity that cannot fit throws
TableEntityTooLargeException. Used by the event, signal, transition, restore
and rules writers; delete-only batches keep their count chunking.
- StorageErrors: one classifier for RequestFailedException (transient 408/429/
5xx, payload-too-large by status or error code, already-exists incl. the
Azurite 400 shape); the two IsAlreadyExistsStatus copies use it.
- Event writer: per-row fallback only for data-shaped rejections; throttling,
outages and oversize batches propagate. Single and batched writer share one
entity builder.
- IngestTelemetryFunction: payload too large -> 413, transient -> 503 with
Retry-After: 5, anything else 500.
- TableStorageChunking.MaxPropertyChars is the one 30,000-char bound.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The five cross-session index tables (SessionsByTerminal, SessionsByStage,
DeadEndsByReason, ClassifierVerdictsByIdLevel, SignalsByKind) were write-only
since M5.d: no endpoint, MCP tool or web page ever read them, while the
dual-write cost a serial queue send per signal/transition on the ingest path,
a 5-13 min full scan of Signals + DecisionTransitions every two hours and five
property-filter scans per session deletion (audit 2026-09-05 F05).
Removed: the tables from TableNames.All (schema hash changes, next cold start
runs a full pass without them), the telemetry-index-reconcile queue name,
producer, envelope, handler, queue function and 2h timer, the cross-partition
Timestamp scans on the signal/transition repositories, the EnableIndexDualWrite
flag (model, config mapping, admin UI, generated wire types), cascade steps
11-15 (later steps renumbered 11-18), the deletion wrappers and the offboarding
wipe entries. DiscriminatorPkProp stays as a step class so manifests written
before this change still resume.
Order of operations: deploy first, then delete the queue pair and the five
tables in the portal (B-88h). Deleting the tables before the deploy 500s every
delete preview/enqueue and poisons offboarding.
Submodules: .claude/commands (import_session.py), internal (D-196, docs, plan).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Audit 2026-09-05 F01/F02. Analyze, VulnerabilityCorrelate, TenantAutoApprove,
IndexReconcile and ImeMsiArchive are now host-managed queue functions under
Functions/Queue bound to the DataStorage connection (identity-based, resolved
by the host): the Flex scale controller wakes an instance for a message, the
host renews the lease and owns the retry/poison ladder. The four cascade
workers with lease/CAS state stay self-polling BackgroundServices; their base
now receives one message per poll so waiting messages can no longer expire
under another worker. Terminal enrollment alerts and hardware-rejection alerts
are enqueued on notification-dispatch (channel ids only, resolved at send
time) instead of fire-and-forget sends the host could drop on scale-in.
Requires the app setting DataStorage__queueServiceUri before deploy.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every non-2xx body in the Functions project now goes through ResponseHelper
(ErrorAsync / BadRequestAsync / NotFoundAsync / ForbiddenAsync /
UnauthorizedAsync / ConflictAsync / InternalServerErrorAsync): 408 sites by a
pattern-exact rewrite of the CreateResponse + WriteAsJsonAsync + return shape,
50 by hand. The eight private BadRequestAsync shadow helpers are gone; the raw
500 catch blocks use InternalServerErrorAsync (sanitized message, correlation
id, MCP hint). 429s keep Retry-After via retryAfterSeconds.
Specialised error bodies where a consumer reads more than the prefix:
AppHomingDeniedResponse (kebab reason codes now Constants.AppHomingReasonCodes,
probe result), TenantConfigPatchFailedResponse (backupId, drift),
QueryBackendLogsErrorResponse (upstreamCode, statusCode, source, hint,
upstream). Twenty-one message-first 200 bodies that the old error tolerance had
let through are typed too (MessageResponse, PreviewWhitelistActionResponse,
SuccessMessageResponse). New codes: PayloadTooLarge, UpstreamError,
UpstreamTimeout, the three trial-gate codes; the agent security path
(SecurityValidationExtensions) writes the envelope with the retry window.
TypedResponseGuardTests: WriteBaseline is empty again — the anonymous
WriteAsJsonAsync ratchet now covers error bodies with zero tolerated sites.
Manifest and generated TypeScript regenerated.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Every error body now leads with { error, code, correlationId }: the generic
ApiErrorResponse (hint?, retryAfterSeconds?, operation?) plus the specialised
McpQuotaExceededResponse (reordered; `message` became `error`),
DelegatedSlotLimitReachedResponse (+correlationId) and the new
SessionDeletionRejectedResponse (replaces the anonymous delete arms). The
correlation id is stamped by the writers — ResponseHelper.ErrorAsync and the
new ApiErrorWriter for the middleware side (auth 401, policy 403/503, rate
limit 429, MCP quota 429, global exception 500) — never by call sites.
exceptionType and rateLimitInfo leave the wire; Constants.ApiErrorCodes is the
code vocabulary and is exported as the manifest section `apiErrorCodes`
(API_ERROR_CODES for web + MCP). CorrelationIdMiddleware names the readable
headers via Access-Control-Expose-Headers.
Guard: TypedResponseGuardTests no longer tolerates anonymous error shapes;
the 458 legacy sites in 149 files are frozen in WriteBaseline to melt down.
New facts: envelope-prefix reflection, ApiErrorWriter body/header pins,
exact-shape pins for the generic envelope and the changed specialised bodies.
Reverses the "error bodies stay anonymous" clause of D-043 (decision entry
follows with the docs stage).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The first cut refused every reader-zone-fallback endpoint. The DeviceSetup line is
written before the agent starts and is backlog on every device, so that silenced
ANALYZE-ESP-001 for all agents without an era anchor (control sessions 2708e6b4 and
18d80f9c came back "not measurable" after the deploy). Two fallback endpoints share the
writer-zone error and cancel it; only a span mixing a fallback endpoint with an anchored
one (or with "now" for an open phase) is unmeasurable — the a7140f98 shape.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
hybrid_login_pending now measures the absence of a real-user desktop after a reboot
(cancelled by the DesktopArrivalDetector or a real JoinInfo user), is never armed on
WhiteGlove devices (resume marker, Part-2 flag, .part1- archive) and reports the
JoinInfo placeholder as a fact only — on a Hybrid join the AD sign-in never replaces it.
New EntraUserAffinityDetector: armed at the real-user desktop on Hybrid devices, the
first IME-TOKEN-SUCCESS line after the desktop posts ime_user_token_acquired; ten
minutes without one post entra_user_affinity_pending with the IME token-failure codes.
ImeLogTracker exposes raw per-line token callbacks past the historic-replay gate;
HelloTracker surfaces User Device Registration 304/305 as device_registration_event.
ANALYZE-ID-004 v2.0.0 keys on those events (event_type not_exists gained a branch),
GATHER-DEVICE-006 follows, GATHER-DEVICE-007/008 collect the Entra operational log.
Constants + MCP catalog gain the five types (D-072), the guard test learns the
multi-line OnTraceEvent shape, gather_* outputs pass the MCP filter gate by prefix.
CMTrace: ImeLogEraPreScan/CmTraceEraTable anchor the backlog era that holds the
bootstrap execution record against the Deployed install marker (0.13-0.36 s in the
field), whole era or nothing, sibling files by local time; origin era-anchored.
Backend phase_duration/app_install_duration refuse a reader-zone-fallback endpoint.
Decisions D-190, D-191. Session a7140f98.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A service principal (idtyp=app, typically behind a federated credential) is admitted only
with the access_as_application application permission and then keys every role table,
binding, audit actor and throttle bucket as app:<client-id>. Caps at resolution: Viewer in a
tenant, DelegatedReader on managed tenants, never a platform role; the roleless catalog tiers
are closed to applications except GET auth/mcp. MCP guard keys on the same principal key.
Members / MCP Users / Delegated Access get a User / Service principal switch; usage views
badge App; McpServicePrincipalFirstSeen ops event and PrincipalKind request dimension.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
A tenant in the owned group of a permanent-Pro tenant is effectively Pro with source "msp":
badge "Pro (MSP)" (also over its own Pro), the Pro entitlement set without the delegation
right (no transitive delegation, no slots), no trial CTA. Nothing is written for that —
TenantConfiguration.ManagedByProTenantId is a load-time projection from a process-wide
index over the Tenant Groups table (owner tier read raw, never through the config service),
so the end of the delegation, the manager losing its permanent tier, or its offboarding
revert automatically. The one write is the retention grace anchor (ProDowngradedUtc),
stamped by ProConferralService from the four paths that end a conferring relationship.
- FeatureEntitlementCatalog.Resolve → EditionResolution { Edition, Source, OwnPro };
Get(resolution) returns ProViaMsp for conferred Pro; the 3-argument ResolveEdition is
gone so every caller resolves from the config (incl. the projection)
- DelegatedAdminService home gate checks the DelegatedAdminAllowed entitlement, not the edition
- TrialExpirySweep reads through the config service: a managed tenant's expired trial emits nothing
- feature-flags + plan response carry editionSource; config/all projects payingCustomer and
managedByProTenantId; shared manifest + web/MCP codegen regenerated
- PayingCustomer: GA-only sales bookkeeping flag on the plan endpoint, no entitlement effect
- web: "Pro (MSP)" label/tooltips, plan section badge, delegated-access hints, tenant
management chips + paying checkbox
Tests: catalog matrix, ManagedTenantProIndexTests (incl. mutual management without recursion,
fail-closed on storage error), ProConferralServiceTests, projection on every config read path,
feature-flags payload, parity facts, serialization (projection never stored), plan transitions,
edition gate, sweep. Backend 5902 green, web tsc/vitest/lint clean, MCP tsc/vitest green.
Concept doc + D-188 in internal (gitlink); customer docs in the docs repo (MCP redeploy needed).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
CodeQL flagged the new denial flow (cs/log-forging #83-#87): header values and the request
path travel from the deny branch into a log line, the ops-events table and the alert
channels. Headers were already stripped of control characters and capped; the path was not,
and Request.Path is percent-DECODED, so %0A arrives as a newline. BuildPrivilegedDenial now
cleans method and path (LogSanitizer) and caps the path at 256 chars, once, before any sink.
Test pins CR/LF stripping and the cap. The remaining alerts are the known false-positive
class (structured ILogger sinks with sanitized discrete properties).
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Live check after the x29 deploy: a KQL syntax error came back as 502, and the MCP sanitizes
every 5xx down to the error code — the syntax message and the forwarded upstream JSON were
invisible, exactly the parity loss the proxy exists to avoid. Upstream 4xx (other than
401/403) now maps to 400 with message, hint and `upstream`; 401/403 (the managed identity's
missing read grant, with a hint naming the role) and store 5xx stay 502. The MCP's structured
4xx renderer appends `upstream` as a fenced block (capped), and keeps sanitizing 5xx.
Tests: status mapping theory, hint texts, upstream rendering on 400 and non-rendering on 502.
Backend 5849, MCP 693 green.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Any signed-in user holds a token the GA-only routes refuse with 403, so the assume-breach
alarm was a push-flood vector: one event per caller AND route per hour, multiplied by
Flex instances. PrivilegedDenialReporter now keys the throttle on the caller alone (a sweep
over the whole GA surface is one event; every attempt still lands in the Warning trace) and
carries a fixed per-instance window budget: 5 events per hour, then ONE Critical storm
marker (same event type, so the same Telegram rule pushes it) and trace-only silence until
the window resets. Worst case per instance and hour: 6 pushes. Global Reader slips count
against the budget too. Tests pin the per-caller throttle, the storm marker and the Reader
accounting; D-185 amended.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
The report ZIP a tenant admin submitted and the preserved diagnostics copy could only be
fetched with a storage key through an az script. They now follow the diagnostics ticket
model: mint under the caller's JWT, download with a short-lived HMAC ticket, bytes proxied
through the backend — a storage SAS never leaves it.
Backend
- DiagnosticsDownloadTicket gains a purpose parameter (default diag-dl-v1, new
report-dl-v1): a ticket for one download surface is rejected on the other (wrong_purpose).
- POST global/session-reports/download-ticket (GlobalReadOrAdmin, same tier as the portal's
download-url): flat-name guard, existence check, best-effort size, ticket bound to the
caller's home tenant; reuses DiagnosticsDownloadTicketResponse (destination SessionReports).
- GET global/session-reports/download (PublicAnonymous, ticket is the sole authority):
per-IP limit + decode via the new TicketDownloadPrelude (shared with the diagnostics route),
streamed from the session-reports container with BlobStorageService.OpenReadAsync.
- Constants.BlobContainers.SessionReports replaces three duplicated literals.
- Pins: catalog completeness/JWT-exempt theories, auth-skip, no-store, purpose tests.
MCP
- get_session_report_download (platform scope) mirrors get_session_diagnostics: kind
(session-report / preserved-diagnostics / diag-files-report), size, download URL, expiry,
instructions; catalog snapshot updated.
Submodules: session-report-debug skill on the tool, fetch-report-zip.sh deleted; plan/todo.
Tests: backend 5839, MCP 683, solution build green.
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Oliver Kieselbach