Technical issues affected 78 of Namecheap’s shared hosting servers on August 26, the company’s second hosting outage this month. The status post opened with “we are currently facing technical issues on the Shared hosting servers,” listed the affected servers under the labels Business, Premium and Server, and added: “Unfortunately, we have no ETA so far.”

Recovery came the same evening. An update at 21:10 UTC reported the servers back up and under observation, and at 22:45 UTC the company declared the issue resolved. What no update did, from the first line to the last, was say what had gone wrong.

Key facts

  • Scale: 78 servers listed in the status post: 18 under Business, 33 under Premium and 27 under Server.
  • Timeline: status post published at 20:25 UTC with no ETA; servers up at 21:10 UTC; resolved at 22:45 UTC.
  • Cause: not stated in the post or in either of its two updates.
  • Context: the second outage to hit Namecheap’s hosting in 13 days, after the August 13 cooling failure in Phoenix.
A Chiller Failure in Phoenix Took Down Namecheap, Liquid Web and phoenixNAP.

Two Hours and Twenty Minutes From Status Post to All Clear

The status post went up at 20:25 UTC, which is 4:25 p.m. on the US East Coast, and named the affected machines individually: eighteen under the label Business, thirty-three under Premium and twenty-seven under Server. Websites hosted on those machines were affected, the company said, without describing how badly or how long the problems had already been running when the post appeared.

Domain Name Wire, which reported the incident while it was open, noted that the post offered no explanation and no estimated time to resolution. The two follow-up updates kept to the same pattern: “The servers are up and running” at 21:10 UTC, and at 22:45 UTC, “the issues have been resolved and the related services are working as expected.” The interval from the first post to the all clear was two hours and twenty minutes; how long the outage ran before the post is not stated.

Fifteen minutes after the all clear, at 23:00 UTC, Namecheap started emergency maintenance on the MySQL service of Premium 293, one of the servers named in the outage list. A separate status post set a one-hour window, said that only MySQL services would be affected and that sites on the server might show related errors, and recorded the work as finished at 00:00 UTC on August 27. Neither status post says whether the maintenance was connected to the outage.

Thirteen Days After Phoenix, With No Cause Named

On August 13, a cooling failure disrupted Namecheap’s hosting, EasyWP, Private Email and some billing operations for hours. The company identified the cause during the incident. In a blog post the next day, chief executive Hillan Klein described it as “a cooling system failure at the RadiusDC: Phoenix data center,” adding that “RadiusDC instructed us to take services offline.”

The August 26 incident is different on exactly that point. Nothing in the post or its two updates names a cause, identifies a data center, or connects the two events. The status post title describes the August 26 incident simply as “temporary technical issues.”

The scope was narrower this time, too. The new post concerns shared hosting servers only, while the Phoenix failure reached across products and kept services offline for several hours.

What the two incidents share is simple: twice in 13 days, Namecheap hosting customers faced service disruption. The difference is in the disclosure. On August 13, Namecheap identified a cooling failure before services had been restored and expanded on it in the chief executive’s blog post the next day. On August 26, the status page closed the incident without naming a cause.