Rates Engine v0.5.0-rc.70

[v0.5.0-rc.70] — 2026-05-22

Fixed

• SSE streaming endpoints poisoned the API latency SLO (#60).
  The HTTPMetrics middleware recorded time.Since(start) into the
  http_request_duration_seconds histogram for every route. An SSE
  handler returns only when the client disconnects, so that
  "duration" is the connection lifetime (minutes-to-hours), not
  request latency. The histogram tops out at 10s, so every closed
  stream landed in the +Inf bucket and pinned p99 at 10000ms —
  burning the latency SLO and firing ratesengine_api_latency_p99_high
  / _p95_high. A latent bug since the first stream endpoint, it
  became material once /v1/ledger/stream (#58) gave every
  status-page viewer a long-lived connection. Fix: the middleware now
  still counts streaming requests in http_requests_total but skips
  the duration observation for /stream routes.
• galexie-archive tip-lag alert false-fired every partition
  cycle (#61). ratesengine_galexie_archive_tip_lag_high (> 5000)
  and _severe (> 50000, a page) predated the 64,000-ledger
  partition model: galexie-archive-fill mirrors only complete
  partitions, so the archive tip structurally lags live by 0–64,000
  ledgers as live works through the current partition. Both
  thresholds sat inside that normal range. Raised _high to
  > 64000 for 90m (a completed partition stayed un-mirrored past
  two fill cycles) and _severe to > 128000 (≥2 partitions
  behind), and corrected the rule's partition-model documentation.