API status
Live status, uptime history, and incident notifications for every customer-facing surface of message.com. The dedicated status site at status.message.com is in build; this page covers the components we monitor and how to subscribe in the meantime.
For real-time updates, follow our community channels. Incident notifications also fire as banner updates on the dashboard at app.message.com.
Monitored components
Each component below is probed every 30 seconds from three regions (US East, US West, EU West). A failure in two of three regions for two consecutive probes triggers an incident.
| Component | Probe target | Notes |
|---|---|---|
| REST API | app.message.com/api/v1 | JSON over HTTPS, workspace-scoped |
| Widget bundle | app.message.com/widget.js | Cloudflare-edge cached, 50KB gzipped |
| Real-time (Socket.io) | wss://app.message.com/socket.io | Two namespaces: /agent and /widget |
| Email inbound | app.message.com/api/v1/email/inbound | Resend forward + custom domain forward |
| Email outbound | app.message.com/api/v1/email/outbound-webhook | Resend delivery webhooks |
| Calls inbound | app.message.com/api/v1/webhooks/calls/inbound | Inbound call webhook over carrier-grade SIP |
| Dashboard | app.message.com | Static export, Cloudflare-proxied |
| AI inference | Internal | Runs on our own infrastructure; your conversations stay in your account and are never sent to a third-party model provider |
Targets
- REST API. 99.95% monthly availability, p95 under 250ms.
- Widget bundle. 99.99% monthly availability (served from Cloudflare edge).
- Real-time. 99.9% monthly availability, p95 event latency under 100ms.
- Email inbound. 99.9% monthly availability, ticket creation within 5 seconds of delivery.
- Calls inbound. 99.95% monthly availability for the webhook handler.
Incident lifecycle
- Alert fires from our probe network or from an on-call escalation.
- We acknowledge within 5 minutes and post an initial status update.
- Updates every 15 minutes until resolution.
- Post-mortem within 5 business days for any incident over 30 minutes or with material customer impact.
Subscribe
Notification channels (the status subdomain ties them all together when it launches):
- Email. Sign up at the future status.message.com.
- RSS. Standard feed at
status.message.com/rss. - Slack. Personal Slack notifications via the upcoming Slack app.
- Webhook. Subscribe a URL to receive incident events.
Uptime history
The status site exposes 90 days of per-component uptime, including degraded windows where we hit SLO but missed our internal targets. We do not retroactively edit history; once an incident is logged, it stays.
Planned maintenance
Planned maintenance windows are announced at least 72 hours in advance. We schedule them between 02:00 and 05:00 UTC by default, when traffic is lowest globally. Customers on the dedicated infrastructure tier get an extra heads-up window and can request a different maintenance time.
Reporting an issue
If you observe an issue that is not yet on the status site, reach us via the channels on the community page. Include the workspace ID, a sample request ID from the response headers, and the timestamp (UTC). The faster we get those, the faster we can triage.