Brief: upstream Cloudflare degradation on Mar 4

For about 90 minutes on Mar 4, 2026 (08:34 - 10:08 UTC), users in some regions saw failed requests to api.swydex.com. Our internal monitoring stayed green — the tier between us and the public was the affected hop.

What happened: Cloudflare regional issues affecting traffic from parts of Asia and the Middle East. Cloudflare's status page acknowledged the incident at 09:01.

What we did: nothing — we're not in front of Cloudflare for our edge. The regional asymmetry meant some of our customers were affected and most weren't. We sent an email to affected accounts explaining the situation as soon as we confirmed the cause was upstream.

What we're changing: we're building a multi-region origin so a single regional outage at the edge doesn't mean inaccessible API for affected users. Q3 work item.

Apologies if you saw degraded performance. No data was lost; webhooks queued and delivered after the incident.