At around 6:40 a.m. PST on Friday, June 16th, we learned from one of our customers that a redirect from a root domain (e.g.,
https://example.com) was failing to redirect to the "www" version of that website (e.g.,
https://www.example.com). Because our internal monitoring tools didn't show any obvious downtime for hosted sites, our customer support team began the routine process of verifying that the DNS setup for that domain was configured properly.
At approximately 9:30 a.m. PST, we received three more support requests indicating that root-domain SSL redirects weren't working when a domain was configured to use a specific IP address (
22.214.171.124) as the sole
A record in the site's hosting configuration. Once our support team learned that this issue wasn't isolated to just one domain, it was immediately escalated to our infrastructure engineering team.
Our engineering team quickly determined that one of the Amazon Web Services (AWS) servers used for SSL redirection was automatically rebooted by AWS due to "degradation of the underlying hardware" that the server was running on. Some time after the automatic reboot, requests to the server started intermittently failing, and then failing outright. The server was quickly replaced with a fresh instance, and redirect requests to that IP address started working again.
Because the server in question didn't crash outright, our health checks and alerts (which notify our 24/7 on-call engineers) unfortunately didn't notify our engineering team right away. We're working on making sure this never happens again, as described below.
Based on what we know so far, this error impacted a handful of sites that had their
A records set to the
126.96.36.199 IP address. In that situation, the following would have happened:
https://example.com) would lead to a connection error
https://www.example.com) would work as expected
For sites that used the recommended configuration of multiple redundant
A records (both
188.8.131.52), we believe the impact was minimal, as most modern browsers have built-in retrying logic.
Out of the thousands of SSL domains hosted on Webflow's infrastructure, we only have direct knowledge of four sites that experienced the redirect failure. However, for the customers that were impacted, we recognize that it was a very frustrating experience.
We're taking the following steps to prevent this issue from happening in the future:
We sincerely apologize for the downtime and inconvenience this issue caused for those affected. We're working hard on making Webflow Hosting an even more reliable place to host your sites. Thank you for your continued support!