About 1:30 p.m. yesterday, Gothamist LLC employees, including the editors who manage this website, noticed that the system on which articles are composed and published was going sluggish and then unresponsive. Next, the sites themselves failed to load, putting many readers out of luck.

For the remainder of the workday, publishing on DCist was a haphazard venture, as the site was functioning one moment, and crumbling the next. But we were hardly the only ones.

The problems stemmed from Northern Virginia, where Amazon maintains a facility that operates a large segment of its cloud storage service, which houses the data used by many of the world’s most popular websites. But according to retail and online storage giant, the issues that hindered Amazon Web Services’ product known as EC2 didn’t crash the cloud. Rather, EC2 was having “degraded performance” affecting a number of servers at the Northern Virginia outpost.

Effectively, DCist wasn’t dead yesterday, but it was barely twitching. In addition to Gothamist LLC, many other popular web companies were impeded by Amazon’s issues, including Reddit, Pinterest, Foursquare, Airbnb and the magazine Fast Company.

An advisory on Amazon’s status page posted about 2 a.m. Tuesday reports that the company has completed recovery on nearly all the affected equipment.

But internal errors are just one thing that can cause Amazon Web Services to have performance issues that wind up impacting wide swaths of the Internet. The derecho thunderstorm that struck the Washington area on June 29 left a part of an Amazon data center in Northern Virginia without primary and backup power. That outage cut users off from popular applications like Instagram and Netflix.