Wires notification and two-factor services into the backend and consumer, adds the authenticated notification feed/preferences endpoints, and exposes the public 2FA login verification endpoint.
Store delayed warmup engagement actions in Postgres and drain them from the consumer so read, important, and star actions survive worker restarts.
Keep foldering and spam rescue immediate while routing delayed actions to the mailbox's current worker at fire time.
Wire warmup health webhooks in the consumer and log skipped warmup actions when a recipient mailbox has no assigned worker.
Run foldering and spam rescue immediately while keeping only low-stakes engagement signals behind the dwell delay.
Add warmup content generation and admin review surfaces, plus mailbox warmup appeal/status APIs.
Track warmup engagement and tampering signals so unsafe mailboxes can be handled by the warmup flow.
Resolve the main-branch conflict in realtime event publishing by keeping both warmup account-health and audit-created events. Renumber the warmup migrations after the current main migration tail.
Make the warmup task reconciler re-check account state and org warmup access before scheduling replacement tasks. Remove pool membership for accounts whose org can no longer use warmup to avoid repeated no-access task churn.
Adds OAuth-backed integration connection management across the API, repository, event dispatch, migrations, docs, and dashboard UI.
Includes realtime invalidation and small dashboard type compatibility fixes needed for the web typecheck gate.
Go CI fails on golangci-lint's gofmt check. Ran gofmt -w against
every file the linter named plus a handful of others that drifted
during the autonomous-fleet work. No semantic changes — alignment
of struct field whitespace and one mis-indented import block.
gofmt -l ./... is now empty; go build + go vet are clean.
cmd/backend/main.go:
- kms.FromEnv replaces kms.New (defaults to AWS, accepts local)
- encryptedkeys.FromEnv with Deps{DB, Dynamo}; default postgres
- codec.NewAvroFromClient wraps the existing Schema Registry client
- eventbus.FromEnv with kafka default; KafkaBus.Producer().WithAvrov2
preserves the existing Avro wire format on Kafka
- events.NewPublisher takes (bus, codec) instead of (producer, avrov2)
- settings.Registrar reflects KMS / EncryptedKeys / Blob / EventBus
choices into storage_backends on boot
- Handler gains EncryptedKeys + StorageBackendRepo for the new admin
and internal endpoints
cmd/consumer/main.go: same eventbus + codec + encryptedkeys swap; the
legacy kafkaProducer is kept around for the consumer's tracking pipeline
which still uses *kafka.Consumer directly (follow-up refactor).
New background job in the consumer process:
1. Pulls up to 1000 mailbox candidates joined with their worst warmup
health state (across all pools they participate in) and their
current worker's risk_pool. Dedicated workers are excluded — single
tenant, segregation not applicable.
2. Recomputes risk_band from health state via RiskBandFromHealth.
If it changed, writes the new band.
3. If the band's matching pool doesn't equal the worker's pool, picks
a new worker via SelectSharedWorkerForBand and migrates the mailbox.
Increments/decrements account counts.
4. Logs each migration to admin_audit_log with action=
"risk_rebalance_migrate" so operators see what moved and why.
Boot-time run + hourly ticker. Rebalancing is intentionally batch, not
event-driven: warmup health states change on a slow rolling-window basis
(warmup_health_sweep is also hourly), so reacting in real time gains
nothing and would cause thundering-herd migrations.
JobsService gets an AssignmentService dep. Nil disables the job (lets
self-hosters opt out by simply not wiring it).
When the dead-worker job reassigns email accounts from a worker whose
heartbeat expired, write a row into admin_audit_log so the dashboard's
audit viewer surfaces these system actions alongside admin-driven ones.
admin_user_id is uuid.Nil (the platform identity), so admins searching
the log can distinguish "system did this" from "an admin did this" by
filtering on that ID. Details include the replacement worker, account
count, and reason.
JobsService gets an optional AdminRepo dep. Nil disables logging — keeps
the contract loose for any other call site that doesn't have one.
Workers heartbeat into Redis every 90s as RFC3339 timestamp values with a
3-min TTL. The dashboard surfaces liveness based on workers.last_seen_at,
but until now nothing populated that column — the "Live" badge was always
red.
New 60s job in the consumer reads each active worker's Redis heartbeat
value, parses the timestamp, and writes it to workers.last_seen_at. Runs
on its own interval (separate from the 5-min dead-worker detection job,
which does heavier reassignment work) so the UI sees fresh data within a
minute.
Matthew Meszaros