Jakarta, domclub Indonesia
—
Cloudflare
said the global network problem that caused many sites and digital platforms to go down on Tuesday (18/11) had been fixed and their services had returned to normal.
“This incident has been fixed. Cloudflare services are currently operating normally. We no longer detect increased errors or latency across the network,” Cloudflare wrote in its status update, Wednesday (19/11) early morning.
“Our engineering team continues to closely monitor the platform and conduct more in-depth investigations into previous outages, but no configuration changes have been made at this time,” he added.
ADVERTISEMENT
SCROLL TO CONTINUE WITH CONTENT
Cloudflare’s network began experiencing serious disruptions delivering core network traffic on Tuesday afternoon.Sites that use Cloudflare in their infrastructure display an error page indicating an interruption in the Cloudflare network.
Cloudflare said this issue was not caused by a cyber attack.Instead, this was triggered by a change to the permissions of one of their database systems, which caused that database to generate duplicate entries into the “feature file” that Cloudflare’s Bot Management system uses.
“The feature file, in turn, doubled in size. The larger than expected feature file was then distributed to all the machines that make up our network,” wrote Matthew Prince, CEO of Cloudflare in a statement on the company’s blog site, Wednesday (19/11).
The software that runs on these machines to direct traffic on the Cloudflare network reads these feature files to keep the Bot Management system up to date with ever-changing threats.
This software is said to have a feature file size limit that is smaller than double the file size.This causes the software to fail to function.
Prince and his team initially suspected the problem was caused by a large-scale DDoS attack, but they later identified the core issue and stopped deploying a feature file that was larger than expected, and replaced it with an earlier version.
“We apologize for the impact this has had on our customers and the Internet in general. Given the importance of Cloudflare in the Internet ecosystem, any disruption to our systems is unacceptable,” Prince said.
“The fact that our network was unable to direct traffic for a period of time was extremely painful for every member of our team. We know we have let you down today,” he added.
A number of sites and platforms that use Cloudflare in their infrastructure experienced downtime on Tuesday evening.
According to monitoring by domclubIndonesia.com at 18.40 WIB, a number of users were unable to access social media X, both on web browsers and mobile applications.
When I wanted to check on the Downdetector platform, this platform was also down.Both give the same notification, namely that there is a problem related to Cloudflare.
Apart from X and Downdetector, the BMKG website and the Perplexity AI platform are also inaccessible and display problem notifications on Cloudflare.
At 19.10 WIB, Downdetector was accessible and showed problem reports on a number of platforms.
More than 1,000 down reports related to Cloudflare have entered the platform.Meanwhile, platform X was reported to be down by more than 700 users.
(lom/fea)
[Gambas:domclub Video]
