November 8, 2007 – (HOSTSEARCH.COM) – Managed hosting services provider NaviSite (http://www.navisite.com/) has reported that 60%-65% of the 150,000-200,000 (depending on which report you read) customer websites affected by its recent outage are now back online. The sites were taken offline by a failed server transition that began last Saturday. Although the company informed its customers their sites would be offline for 12 hours, many sites were in fact offline for around 5 days and more.
The problem started with NaviSite’s plans to replace some of the servers located at the premises of its recent acquisition, Alabanza. The plan was to move 200 Alabanza servers from the company’s offices in Baltimore to NaviSite's Andover datacenter. After physically moving the servers, the intention was then to ‘virtually’ migrate data from older servers to the new machines. The migration experienced Address Resolution Protocol issues causing synchronization failures, and as a result, the virtual transition took longer than expected. To speed up the process, and meet its suggested schedule, the company decided to physically move another 200 servers from Baltimore to Andover by truck. Once installed, it became clear that there were IP address issues, with customers not being able to access their sites through their URLs. When other customers became aware of the situation, they went online, stretching the company’s network, and ultimately causing it to overload.
Although many sites are up, it is still unclear how long other sites will remain impacted. A number of the company’s customers have publicly expressed their dissatisfaction with the way the issue was managed, with some asking for government sanctions against the company. "We take it very seriously. This impacted so many people and so many businesses. We're doing it 24/7. This company is trying to get this thing fixed," NaviSite's Chief Marketing Officer Rathin Sinha was reported as saying.
