Fastly's Compute and Next-Gen WAF (NGWAF) services returned elevated errors for 37 minutes early on 28 October 2026. The fault sat on Fastly's side and was confirmed mitigated by their own engineers.
Started
00:10 UTC
Duration
Lasted 37m
Source
IsDown
Next time
Fastly is degraded. Is your product working?
An incident on their side does not always mean an outage on yours, and the only way to know is to already be watching your own endpoints. Set that up in under a minute and you will be notified the next time your own site is affected. And there will be a next time.
50 monitors for your own websites. No card.
Affected components: AI Accelerator, Config Store, Edge Cloud Services, Fanout, Fastly Compute, KV Store, Next-Gen WAF (NGWAF), Secret Store, Websockets
What happened?
Fastly opened the incident at 00:10 UTC, reporting elevated errors to Compute and Next-Gen WAF (NGWAF) services. At 00:34 their engineers said they had identified a contributing factor and were developing a fix, and added that all other products and services were unaffected. By 00:35, a minute later, they confirmed the impact had been mitigated, and the incident closed with a total duration of 37 minutes. Nine components appear on the affected list for the day, including Fanout, Config Store, KV Store, Secret Store, AI Accelerator, Edge Cloud Services, Websockets and the two named above, though Fastly's own text narrows the actual impact to just Compute and NGWAF. No cause is named beyond "a contributing factor", and Fastly has not published anything further explaining what that factor was. No user reports were filed during the window, so there is no independent account of what this looked like from outside Fastly's network. Three updates were posted in total and all three are shown here. Over the last 90 days Fastly has logged 19 incidents with a median duration of 6 hours 44 minutes, which makes this one short by their own recent standard. Across all time Fastly has recorded 1,007 incidents at a rate of 13.1 per month, so isolated short-lived errors like this one are a routine part of running on their edge network.
Learning
UptimeRobot's own probes against Fastly's api-control component showed 100% uptime across 24 hours, 7 days and 30 days, and this incident was not registered against that check. That is not a contradiction. Fastly's problem was isolated to specific services, Compute and NGWAF, and a probe watching a different component has no reason to see it. This is the general shape of provider-side failures: the breakage can live inside one service while the rest of the platform, and any monitoring aimed at the wrong corner of it, stays green. Over the last 30 days UptimeRobot recorded 6 incidents against Fastly, all resolved, with a median duration of 52 minutes, a separate and useful baseline from Fastly's own 90-day median of nearly 7 hours. If your product runs on Fastly Compute or the WAF and you only watch your own endpoints, an incident like this can pass through your infrastructure without ever showing up in your dashboards. Your monitoring covers your half, UptimeRobot now watches the provider's half too, so next time you know which side broke without guessing.