Rackspace suffered another data centre failure that caused outages for its cloud computing customers, the third such incident since late June.
Power failures hit Rackspace's 144,000-square-foot data centre in the Dallas suburbs on June 29 and July 7, affecting about 2,000 customers. Another outage hit the same facility this week.
Rackspace was performing scheduled maintenance on a UPS cluster "when a short occurred and caused us to lose the PDUs [power distribution units] behind this cluster," Rackspace reported in its official blog at 3:27am Rackspace aborted the maintenance for the rest of the day and rescheduled it for another date.
This week's outage affected Rackspace's Cloud Sites, a website hosting service, and Slicehost, which provides hosted virtual servers. The power distribution units were down for only five minutes, but the incident apparently caused problems for several hours. Slicehost was still reporting "intermittent issues" at 7:09am, and degraded performance issues were affecting Cloud Sites until 9:08am.
In a Twitter post, one Rackspace user reported that his websites went down for seven hours due to the Dallas outage.
After outages in June and July, Rackspace was forced to pay out between $2.5 million and $3.5 million in service credits to customers. At the time, Rackspace CEO Lanham Napier acknowledged the outages harmed the company's reputation.
"Any time we have an incident like this, it does impact our credibility," Napier said in July. "The only way we earn it back is we have to execute at a high level for a long time."