The company Cloudflare, which provides essential services to many websites, experienced some problems on June 21. The situation is resolved, but you may have had problems browsing online.
It’s a bit of a rough June 21st for Cloudflare, and maybe for you too. The American company has just reported in the morning disruptions in its infrastructure, which are likely to degrade the web browsing of multiple Internet users. But the situation is being restored, according to the most recent progress points from Cloudflare.
” The Cloudflare team is aware of current service issues and is working to resolve them as quickly as possible. », was writing early in the morning Cloudflare’s official Twitter account. “The problem has been identified and a fix is being implemented,” then announced a second message published about twenty minutes later.
Cloudflare, an internet giant that acts behind the scenes
Cloudflare is a company that is generally unknown to the general public, but whose behind-the-scenes role has become very important in keeping the web running smoothly. The company provides all kinds of services to sites (for example tools to resist DDOS attacks, to cache content and facilitate its distribution on the net or to direct navigation with the DNS).
The malfunction reported by Cloudflare is significant. ” Cloudflare’s network connectivity was disrupted in large areas “, according to the dashboard dedicated to the state of health of its systems. ” Customers attempting to reach Cloudflare sites in affected regions will see 500 errors. The incident impacts all data plane services in our network. »
500 errors are codes that indicate a problem with the server. There are others, some of which just indicate that everything is fine (a 200 code means that everything is fine). Some are very well known: a 404 code designates a resource that cannot be found, such as a web page. These HTTP codes are numerous. Cloudflare also uses several for itself.
Among the services victimized by Cloudflare’s temporary setbacks are Discord (a VoIP platform), Omegle (an instant messaging and webcam chat service), NordVPN (a VPN provider), Crunchyroll (SVOD specializing in animation Japan), DoorDash (food delivery), Feedly (RSS feed aggregator), Coinbase (a cryptocurrency exchange, gaming League of LegendsShopify (e-commerce), but also media like Register and Medium, reports Techcrunch.
Cloudflare’s DNS resolver has also been acting up
Also disabling for browsing the net, the incident impacted Cloudflare’s public resolver, observes Stéphane Bortzmeyer, R&D engineer at AFNIC, the organization that manages the top-level domain name assigned to France (“.fr”). Cloudflare provides DNS servers (184.108.40.206 / 220.127.116.11) to route site access requests.
To put it simply, the DNS is a registry that knows how to match the name of a website with the IP address of the server on which it is hosted. The web browser queries the DNS resolver to tell it in a way: what is the path to take to get to this site? The resolver consults its database and returns the response with the IP address.
As a general rule, Internet users use by default, and without knowing it, the DNS of their Internet service provider. These are usually sufficient for the vast majority of web activity, but there are times when they run into trouble — and when a DNS resolver fails, it can quickly feel like the whole web is down. broke his face.
Many companies offer public DNS, including Google, Cisco (with OpenDNS), or Verisign. There are others, such as OpenNIC and FDN. And the European authorities want to have their own DNS resolver, for reasons of sovereignty. It can be useful to know how to change your DNS so that you don’t have to depend on the intervention of the organization that provides them.
It might not be necessary to make the switch this time, because according to Cloudflare, everything is back to normal. ” A fix has been implemented and we are monitoring the results another tweet read. ” We are still monitoring the outcome of the correction put in place “added the company, before concluding:” This incident has been resolved. »