(WEB HOST INDUSTRY REVIEW) — Early Tuesday morning, Rackspace Cloud (www.rackspacecloud.com) engineers saw intermittent connectivity problems with the WC2 cluster at the web hosting provider’s Dallas-Fort Worth data center due to intermittent power connectivity following uninterruptible power supply maintenance.
The majority of service had been restored to the WC2 cluster within an hour after Rackspace first noticed the issue at 12:35am CST, however, some sites still had performance issues, mainly dealing with SSL clusters. After two hours, Rackspace reported that all services and clusters were online and responding for Cloud Sites in WC2, but there were still a few residual DNS and FTP cluster pool issues in WC2, which were dealt with by 4:30am, restoring service to normal.
SSL issues resurfaced at about 6am and were promptly dealt with by Rackspace Cloud engineers. At 9am, Rackspace noticed slow performance on sites using the PHP 5-3 Cluster, which were also quickly corrected.
“Rackspace has experienced a service interruption during tonight’s scheduled maintenance on UPS Cluster G,” stated Debbie Talley in a company blog post. “We were testing phase rotation on a Power Distribution Unit when a short occurred and caused us to lose the PDUs behind this cluster. The phase rotation allows us to verify synchronization of power between primary and secondary sources.”
The power distribution units were only offline for a total of five minutes, and their maintenance was rescheduled to deal with the more pressing matter of the intermittent Cloud Sites downtime.
Just one of its nine worldwide facilities, Rackspace’s Dallas-Fort Worth data center was hit by two notable outages over the summer. In August, part of the Dallas-Fort Worth data center went offline for approximately 45 minutes. A week later, a bus duct failure caused customers to lose network access for a short period of time.
Rackspace guarantees 100 percent network and infrastructure uptime in its service level agreement. In terms of compensation, the SLA is fairly straightforward: “Rackspace will credit your account 5% of the monthly fee for each 30 minutes of infrastructure downtime, up to 100% of your monthly fee for the affected server(s).”
