# Canvas by Instructure outage on 21 October 2026

> On 21 October, Canvas by Instructure had a major outage in which users of Canvas Data 2 received 500 errors. The problem lasted about 1 hour 50 minutes, from 21:07 to 22:57 UTC; the status page listed both Canvas Data 2 and its AWS eu-west-1 region as in major outage from 21:06 to 22:54. According to the status page, the errors were due to a vendor outage, and a vendor fix was being monitored from 21:56 before the incident was marked resolved at 22:54. Claude, 20 Minutes and Musixmatch also had problems at the same time.

- Service: [Canvas by Instructure](https://stackpulse.app/services/canvas-by-instructure.md)
- Severity: major_outage
- Title on the status page: 500 errors when accessing Canvas Data 2
- Started: 2026-10-21 21:07 UTC
- Ended: 2026-10-21 22:57 UTC
- Lasted: 1 h 50 min
- Incident open on the status page: 21:06–22:54 UTC
- The status page's incident: https://stspg.io/nvhltx1jtwsp
- Page: https://stackpulse.app/services/canvas-by-instructure/outages/2026-10-21-2107

## What broke when (UTC)

| Component | Status | From | To |
|---|---|---|---|
| Canvas - Canvas Data 2 | partial outage | 2026-10-21 21:06 | 2026-10-21 22:54 |
| AWS Region - eu-west-1 | partial outage | 2026-10-21 21:06 | 2026-10-21 22:54 |

## At the same time

- [Claude](https://stackpulse.app/services/claude.md): issue from 2026-10-21 20:32 UTC
- [20 Minutes](https://stackpulse.app/services/20-minutes.md): major_outage from 2026-10-21 21:31 UTC
- [Musixmatch](https://stackpulse.app/services/musixmatch.md): issue from 2026-10-21 21:41 UTC

Problems at the same time are a clue, never proof that one caused another.

Stackpulse isn't affiliated with Canvas by Instructure.