Salesforce.com Inc.’s crash just days before Christmas came amid a spate of outages at hosted service providers, a run of bad luck that suggests the “on-demand” software wave may require customers to temper their demands.
The company, which provides on-demand customer relationship management (CRM) tools, has been tight-lipped about the roots of the outage. In a statement, the company said some users experienced intermittent access over an approximately six-hour period Dec. 20. on one of four global nodes, caused by an error in their database.
Anecdotal reports suggest that the breakdown was among the most severe in Salesforce.com’s six-year history.
“We are told the outage impacted a majority of customers, as well as restricting the company’s own access to the system,” First Albany Corp. analyst Mark Murphy wrote in a research note. “We are aware of multiple customers that are quite displeased with the outage.”
In October, Salesforce.com expanded its presence in the Canadian market, opening an office in Toronto that can support up to 150 people, led by vice president Michael Basch. Basch declined to comment on the outage’s impact on their Canadian customers, which include Bell Canada, Nova Scotia Business and Blast Radius.
When contacted by ComputerWorld Canada Bell Canada declined to comment on the outage, while Nova Scotia Business and Blast Radius said they had not been impacted.
However, in the U.S. user angst is simmering. “Click and wait, click and wait,” was how customer Charlie Crystle described his experience. The CEO of nonprofit fund-raising software maker Mission Research said although full-blown outages are rare, he estimated that at least once a week the system slows to the point of being nearly unusable.
“I’ve got a sales team that ends up really unhappy. They want to get in there and close sales,” Crystle said. He said he plans to soon cancel his Salesforce.com licenses and switch to a homegrown, in-house sales management system.
Salesforce.com’s blackout followed similar downtime from other “software as a service” providers. Six Apart Ltd.’s TypePad blog hosting service went down for a day in December following a failed storage upgrade. Affected customers included Major League Baseball’s MLB.com site, which hosts all of its blogs with TypePad. In addition, the del.icio.us bookmark-sharing service that Yahoo Inc. just bought suffered days of problems after its data centre lost power.
A competitor to SalesForce.com in the hosted CRM space, San Mateo, Calif.-based NetSuite Inc. said it was surprised to learn of the extent of the SalesForce.com outage, but doesn’t think it will shake confidence in the growing market space.
“We were fairly surprised that it was such a substantial outage, because the online hosted application industry is maturing and the service has been fairly reliable,” said NetSuite regional vice president Marc Huffman, who heads the company’s Canadian office in Mississauga, Ont.
Huffman wouldn’t say if he’d heard from Salesforce.com customers looking to switch service providers because of the outage, but he did say it would be natural after such an occurrence for companies to examine their alternatives. He said one thing NetSuite did learn was that its 99.9 per cent contractual uptime guarantee appears to be fairly unique in the market.
“Quite honestly, I think we’re going to start to market that a little more heavily,” said Huffman.
Yankee Group Inc. analyst Sheryl Kingstone said she hopes occasional problems such as Salesforce.com’s stumble won’t scare customers away from on-demand services. Internal software is hardly immune to breakdowns, she said. “I wouldn’t say this is indicative of a trend, or likely to happen again,” Kingstone said of Salesforce.com’s outage. “A lot of the vendors are putting very large investments in place to prevent these sorts of situations.”
Salesforce.com, based in San Francisco, Calif., is in the midst of a major update and infrastructure overhaul. It’s sinking $50 million into setting up new data centres in the U.S. East and West coasts with redundancy and fail over capabilities, an initiative it calls “Mirrorforce.”