We are currently experiencing intermittent network connectivity issues that may affect website, email, SQL database, and other hosted services. Our investigation indicates that the issue is related to a broader outage affecting Cogent Communications, a major global Tier 1 Internet backbone provider that carries traffic between Internet networks. Reports indicate that Cogent is dealing with multiple fiber cuts affecting the Los Angeles and Southwestern U.S. region. Cogent engineers and their fiber providers are currently working on repairs. At this time, no estimated time of resolution has been provided. Additional reports and discussion regarding the Cogent outage can be found here: https://www.reddit.com/r/sysadmin/comments/1w5hszn/cogent_outage/ We will continue monitoring the situation and provide further updates as they become available. We apologize for the inconvenience.
We have determined that the intermittent connectivity issue is related to a broader Internet routing/peering problem involving Cogent. To clarify, Cogent is not one of our upstream Internet providers, and our network does not depend on Cogent for connectivity. Our infrastructure uses multiple redundant upstream providers, including Lumen and Zayo, which operate independently of Cogent. This redundancy is specifically designed so that the failure of one provider or route does not take our network offline. However, Internet traffic does not always travel directly from a customer’s ISP to our network. Depending on the customer’s location and ISP, their traffic may pass through several intermediate backbone networks before reaching Lumen or Zayo. In some cases, that path may include Cogent or an interconnection between Cogent and another carrier. Therefore, a routing or peering problem involving Cogent can affect only certain paths across the Internet, causing some customers to experience intermittent website, email, or SQL connectivity while other customers can reach the same services normally. Our servers and redundant upstream connectivity remain operational. The issue is occurring outside of our network on certain Internet routes, and we will continue monitoring the situation as the affected backbone routes stabilize.
