CloudflareARN (Stockholm) on 2026-11-11
The scheduled maintenance has been completed.
Resolved
Outages and recoveries from the services you depend on.
40 incidents
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
This issue has been resolved, and the miscounted views have been removed from the Analytics dashboard. Agent views and total Views now show corrected numbers, which will be lower than before because they were previously overstated. Some sites may also notice smaller corrections in earlier months. If you exported analytics recently, export them again to get the updated figures. Human visitor metrics were not affected.
Resolved
This incident has been resolved.
Resolved
The scheduled maintenance has been completed.
Resolved
Responses took longer than usual to start. Pay-as-you-go: deepseek-v4.1-flash experienced slower responses.
Resolved
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Maintenance
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
The maintenance is complete.
Resolved
This incident has been resolved, and service is operating normally again. Thanks for your patience.
Resolved
Responses took longer than usual to start. Free inference: beam-free experienced slower responses.
Resolved
The maintenance is complete.
Resolved
The scheduled maintenance has been completed.
Resolved
Responses took longer than usual to start. Free inference: mimo-v2.6-flash-free experienced slower responses. Go: deepseek-v4.1-flash experienced increased error rates.
Resolved
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
This issue is resolved. If you are still seeing issues, please contact support.
Resolved
This incident has been resolved.
Resolved
Responses took longer than usual to start. Pay-as-you-go: deepseek-v4.1-flash experienced slower responses.
Resolved
Service has recovered. Error rate: 0.00%
Resolved
Issue resolved
Resolved
Issue resolved
Resolved
Responses took longer than usual to start. Free inference: deepseek-v4.1-flash-free experienced slower responses.
Resolved
An upstream model provider had problems, so some requests failed. Go: deepseek-v4.1-flash experienced increased error rates.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
This incident has been resolved.
Resolved
The issue affecting Notion Agent sessions has been resolved
Resolved
We have resolved the issue causing ingestion latency for replay events in our EU region.
Resolved
The incident has been resolved and SMS delivery receipts from Twilio Phone Numbers to Movistar network subscribers in Venezuela are operating normally.
Resolved
The incident has been resolved and SMS delivery receipts from Twilio Phone Numbers to Movistar network subscribers in Venezuela are operating normally.
Resolved
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
The scheduled maintenance has been completed.
Resolved
Front has submitted new app versions to both app stores and is awaiting their review. We anticipate the latest fixed version will be available in a matter of days. In the meantime, see above for possible workarounds.
Resolved
What happened At 9:00 AM ET on November 10 we shipped a change to how our servers cache data. A side effect was that our servers began requesting font data from Google far more often than before. Google caps how many of these requests we can make per day. That afternoon we hit that cap, and Google began rejecting our requests. Page loads that depended on that font data stalled or failed, and the failed requests were retried, which kept us pinned at the limit. Timeline (November 10, ET) - 2:10 PM: Google begins rejecting our font requests. Errors and slow page loads start. - 2:20 PM: First user reports appear on the forum. - 2:26 PM: Our monitoring alerts fire and engineers are paged. - 2:39 PM: We trace the problem to the 9:00 AM change and decide to roll it back. - 2:49 PM: We confirm the usage cap is the cause, ask Google for a higher limit, and begin provisioning separate capacity so we don't have to wait for that request. - 3:00 PM: The rollback is live. It does not restore service on its own, because we are still over the cap. - 3:08 PM: We find that our servers have not picked up the new capacity. We correct this and restart them. - 3:09 PM: Uptime checks pass again. - 3:15 PM: Errors have dropped off and we move to monitoring. What we're doing about it - The change that caused the increase has been rolled back. - We have provisioned additional capacity and asked Google for a higher limit. - We are changing how fonts are loaded so that page loads no longer depend on a request to Google. This work is in progress.
Resolved
The scheduled maintenance has been completed.
Resolved