Staff of the large web-solutions provider Endurance International Group traced Friday's day-long outage of the company's data center in Provo, Utah, to two core switches in its network.

Ron LaSalvia, the company's COO, wrote in a status update that a problem occurred during routine server maintenance. A hardware failure “cascaded” across the network, setting off the outage that started in the morning on 2 August, impacting operations of the firm's hosting customers.

“Our data center in Provo, Utah, experienced an outage that impacted many of our Bluehost, HostGator, HostMonster and Just Host customers, causing intermittent connectivity to your sites and our internal support and phone systems throughout the day,” LaSalvia wrote. These are four of the brands the company provides its web solutions under.

The company's response team, which included customer service teams, technicians and engineers, worked to restore services out of its command center in Burlington, Massachusetts. The team provided updates on the recovery effort every 30 minutes on a website that was set up specifically to communicate progress of the effort.

The two core switches were identified as culprits by about 3:30pm EST. One of the switches had been restored by the time of the 4pm status update.

All customers were back online by 8pm.

“We understand that your sites are your lifeblood, as well as the engine to our economy, and we have committed all company resources, nationwide to a swift resolution and full restoration,” Salvia wrote in one of the updates.