Live  /  Amp  /  Incident

Amp: Agents fail to start for some threads

Amp · ampcode.comMajor incidentResolved2026-10-01 04:00 UTC
Public — what happened

Summary

Amp reported a major incident beginning at 03:27 UTC on October 1, 2026, in which agents in a portion of threads could not start or reconnect, though the thread data itself kept loading. The company said it suspected elevated CPU load on certain nodes during a deployment was behind the problem. By 03:49 UTC Amp said recovery was widespread, with only a small number of threads affected and agents unavailable for at most a few minutes, though some CPU readings were still higher than desired. The incident was marked resolved at 04:00.

Timeline

Source of record: provider status page. Summary written by SquidGrid from the provider’s published updates; times in UTC.

Independent path health

The status page is the symptom. SquidGrid measures the paths that reach Amp against their own baselines — whether the routes carrying your traffic degraded, at which handoff, and who owns that stretch.

🔒 Sign in to see
Customers get the live Provider Assessment Report for Amp: its full incident track record, what tends to break, and recent public reports of problems.

Included with any paid plan.

Open the Amp report →

More from Amp

Covered in The Squid’s Ink, Week 41, 2026, our weekly internet health report.

The full archive is on the Amp status & history page.

Related: Amp status & history. SquidGrid is not affiliated with Amp.