# What broke on 2026-08-20

- Incidents: 93 at 60 providers; major outages: 4
- Downtime: 3d 13h

## Worst
- **Google Cloud**: [We are investigating an issue where customers may experience timeouts, service degradations, errors, and elevated latencies across multiple products in the us-west1 region.](https://down.guide/gcp/DWN-dfad8e7abb) (Major outage; lasted 3h 40m; started 2026-08-20 15:40 UTC; AlloyDB for PostgreSQL, Apigee Edge Public Cloud, Apigee X +24; global, us, us-west1)
- **Anthropic**: [Elevated errors on requests to multiple models](https://down.guide/anthropic/DWN-cae09f0754) (Partial outage; lasted 26m; started 2026-08-20 19:16 UTC; claude.ai, Claude API (api.anthropic.com), Claude Code +1)
- **Confluent Cloud**: [Elevated error rates and latency in GCP us-west1](https://down.guide/confluent/DWN-7712c29e92) (Partial outage; lasted 3h 8m; started 2026-08-20 17:10 UTC; Confluent Cloud; us, us-west1)
- **GitHub**: [Intermittent failures creating agent tasks](https://down.guide/github/DWN-0ca901ec02) (Major outage; lasted 9h 54m; started 2026-08-20 14:43 UTC; us)
- **Anthropic**: [Elevated errors on Google connectors](https://down.guide/anthropic/DWN-16402c4e99) (Degraded; lasted 29m; started 2026-08-20 18:32 UTC; claude.ai, Claude API (api.anthropic.com), Claude Code +1)
- **Ably**: [Elevated Error Rate in ap-southeast-1](https://down.guide/ably/DWN-6d6a7c3163) (Partial outage; lasted 3m; started 2026-08-20 19:07 UTC; Asia-Pacific (Singapore); ap-southeast-1, apac)
- **Cloudflare Workers**: [Durable Objects and Downstream Service Errors](https://down.guide/cloudflare-workers/DWN-8e6cbc35e8) (Degraded; lasted 1h; started 2026-08-20 16:48 UTC; Durable Objects, Artifacts, AI Search +4)
- **Better Stack**: [Degraded Telemetry performance](https://down.guide/betterstack/DWN-2ede57a10a) (Major outage; lasted 9m; started 2026-08-20 14:39 UTC; us)
- **Akamai**: [Reporting data availability issues in 'Traffic by Hostname' report](https://down.guide/akamai/DWN-6abffd724b) (Degraded; lasted 8d 5h; started 2026-08-20 16:43 UTC; Reporting; us)
- **Harness**: [All modules are running slow in Prod1/2/3/4 due to cloud provider incident](https://down.guide/harness/DWN-1a048fd04b) (Partial outage; lasted 2h 7m; started 2026-08-20 17:17 UTC; Platform, Platform, Platform +1; us, us-west1)

---
down.guide reads providers' official status pages every minute and adds its own evidence, always labeled. Free, no account. Agents: MCP at https://down.guide/mcp, REST at https://down.guide/api/v1, feeds at https://down.guide/feed. Built by Sazabi (https://sazabi.com/?utm_source=down-guide&utm_medium=markdown).