Increased errors for R2 and Tiered Cache

Summary

The provider says a network connectivity issue between several data centers from 05:00 UTC caused some R2 and Tiered Cache requests to fail, most in Chicago and Newark. Error rates are near normal since 05:30 UTC, and monitoring continues while the network issue is resolved.

Written by anthropic/claude-haiku-5.5 from Replicate's own updates.

What to do

• Check whether you use Replicate Chicago, United States - (ORD), R2, Newark, United States - (EWR) in us. Check your stack

  Drop a package.json, lockfile, Terraform or a list of domains at down.guide to see every provider you depend on.

• Check that everything you run on Replicate came back after the window closed 2026-11-14 06:45 UTC.

• Subscribe to Replicate's feed to hear about the next one first. RSS feed

  https://down.guide/feed?provider=replicate works in Slack (/feed subscribe), Discord feed bots and any RSS reader.

Updates from Replicate 3

1. investigatingNov 14, 2026, 5:30 AM UTC

   Cloudflare is aware of, and investigating an issue with R2. Customers in Chicago, IL, may see 503 errors when using R2. We are currently investigating this issue.

   provider

2. monitoringNov 14, 2026, 5:43 AM UTC

   Starting at 05:00 UTC, a network connectivity issue between several Cloudflare data centers caused a small percentage of requests to fail. R2 customers may have seen HTTP 500 errors, and requests for content served through Tiered Cache may have returned 520, 522 or 524 errors. Customers in Chicago and Newark saw the highest error rates, and other locations were affected at lower rates. Error rates have dropped close to normal levels since 05:30 UTC. We are continuing to monitor while the underlying network issue is resolved.

   provider

3. resolvedNov 14, 2026, 6:45 AM UTC

   This incident has been resolved.

   provider

Corrections

None so far.

Something wrong on this page? Report an error or send us a note. Every fix is listed here.