A Post Mortem on this Morning's Incident
On June 17 and June 20, Cloudflare experienced internet outages due to significant packet loss on one of their major transit provider backbone networks, Telia Carrier. The company's systems detected the problem instantly and recorded it. They identified weaknesses in communication during these incidents and are taking steps to improve their response time and accuracy. Cloudflare is working towards building a resilient network through greater interconnection, automated mitigation, and increased failover capacity. They plan to extend their proactive packet loss detection mechanism to all their POPs within the next two weeks.
Company
Cloudflare
Date published
June 21, 2016
Author(s)
Jérôme Fleury
Word count
1010
Hacker News points
None found.
Language
English