Status
Public web, documentation, and Workspace API surfaces are monitored separately from private project materials. This dashboard reflects shared infrastructure only.
Customer-facing components across the public platform.
| Component | Description | Status | Uptime (90d) |
|---|---|---|---|
| Website | Public site and marketing pages | Operational | 100% |
| Documentation | Public docs, runbooks, and reference | Operational | 99.99% |
| Workspace API | Primary application API edge | Operational | 99.98% |
| Batch processing | Async job execution (OCR, transcode) | Operational | 99.95% |
| Artifact storage | Durable object store for outputs | Operational | 99.99% |
| Upload intake | Inbound file ingestion and validation | Operational | 99.97% |
| Export pipeline | Document and media export and packaging | Degraded | 99.91% |
| Monitoring probes | Synthetic checks and alert routing | Operational | 100% |
Most recent first.
A subset of export jobs experienced longer-than-normal queue times following a surge in concurrent packaging requests. Additional workers were provisioned and queues are draining; we are monitoring throughput before closing.
Background compaction and index rebuild on the global artifact store completed within the planned window. No downtime; brief read-latency increases recovered automatically.
Rolled out an updated synthetic probe configuration across the platform, adding finer-grained reachability checks for the upload intake path. No customer-facing impact.
A transient dependency timeout caused a portion of OCR jobs to enter retry. Automatic retries succeeded and all affected jobs completed; the underlying timeout was patched.
Routine rotation of internal service signing keys as part of standard credential hygiene. The rotation was transparent to all components.
A malformed-input edge case caused intermittent 5xx responses on the intake path. Input validation was hardened and additional probes were added to catch the pattern earlier.
Conducted a planned failover exercise for the Workspace API edge to validate recovery procedures. Traffic was served throughout with no measurable impact.
Upcoming windows on shared infrastructure.
Rolling node upgrade on the artifact store. No downtime expected; reads and writes remain available, with possible brief latency increases on large objects.
Planned rollout of an updated API edge build. Requests will be drained gracefully between nodes; no interruption expected.