Case · 2026-07-30

Two network failures, one status page, zero limit reset

Network karaoke: 'stable right now' is the new encore.

claudeX @ClaudeDevssev 4/5official
outageinfrareliabilitycapacity

Summary: @ClaudeDevs reported two incidents in 24 hours with elevated errors or reduced availability. Multiple separate network failures cut capacity; some requests failed during traffic reroute. Capacity partially restored, not fully clear. Status page logged multi-hour elevated errors on Jul 29 and again on Jul 30 across many models.

Source: @ClaudeDevs · 30 Jul 2026

What landed

@ClaudeDevs posted that two incidents over the prior 24 hours produced elevated errors or reduced availability for some users. Cause named: multiple separate network failures that cut capacity to serve Claude. Some requests failed while traffic was rerouted.

Follow-up in the same thread: progress restoring capacity, things stable right now, but not fully in the clear. Further updates pointed at status.claude.com.

Corroborating official status incidents in that window:

  • 29 Jul 2026: elevated errors across all models roughly 19:45–21:26 UTC (about 1 hour 40 minutes of elevated rates), then resolved.
  • 30 Jul 2026: elevated errors across many models from about 05:57 UTC, with model-by-model recovery wobbles (Opus, Sonnet, Fable) through late morning UTC before resolve. A separate Opus 4.8 degradation was also logged that day.

Why it matters

When capacity is a network reroute story, users still just see 529s, hangs, and half-finished Code sessions. Official silence until evening, then a dual-incident summary, is product trust theater with receipts.

Satire margin

Two incidents. Multiple separate network failures. Stable-ish. Not fully clear. We clap. We archive. We do not reset your weekly limit for you.

Primary sources

← Timeline· All cases· See a sibling failure? Submit it