Today, starting at approximately 10:03 UTC, a misconfiguration at a major external network provider caused connectivity disruptions across the Internet. These disruptions impacted the operations of many service providers, including Cloudflare. Cloudflare’s systems routed around the problem, however, some Internet properties that use Cloudflare saw increased error rates during the impact period despite our mitigations.During this period we saw traffic drop by about 3.5% across our network as well as an increase in HTTP errors which may have presented as 522 (Connection Timeout) status codes.
Cloudflare have posted an initial detailed analysis of this incident written by Cloudflare CEO on the Cloudflare blog. Presently, Cloudflare is no longer seeing additional impact from these transit provider issues.
These events can be disruptive, Cloudflare will continue to provide the latest updates for issues impacting Cloudflare's network on cloudflarestatus.com.
Below is the sequence of the investigation carried out and the resolution of the problem.
Aug 30, 16:12 UTC
Aug 30, 15:31 UTC
Aug 30, 15:16 UTC
Aug 30, 15:05 UTC
Aug 30, 14:46 UTC
We are applying corrective action in our data centers as the situation changes in order to improve reachability
Aug 30, 14:26 UTC
Aug 30, 13:58 UTC
Aug 30, 12:59 UTC
This is affecting all data centers that make use of this transit provider and we are working on implementing mitigations to alleviate this issue.
Aug 30, 11:57 UTC
Aug 30, 11:39 UTC
Aug 30, 11:08 UTC
We are working to understand the full impact and mitigate this problem. More updates to follow shortly.
Aug 30, 10:45 UTC
Aug 30, 10:21 UTC
Monday, August 31, 2020