On-demand software vendor and Wall Street darling Salesforce.com took a hit on Friday just one day after the company acknowledged yet another service interruption which affected customers in North America.
According to the company's monitoring service, trust .salesforce.com, the interruption on Thursday which affected its North American operations was du to "a software patch affecting the cache server."
According to Salesforce, "the service was restored to full operations after removing the software patch, resetting the cache system configuration, and re-starting the cache servers. Changes to operational procedures have been made to protect the service from similar issues going forward."
While that's nice to know, it's probably cold comfort for the company's customers, many of whom were unable to log in during the outtage. It's bad news, too, for investors: the market is willing to believe that Salesforce's "software as a service" or SaaS model is the future of enterprise computing -- as long as they can keep their service running and available. The company experienced another serious outtage in December, then again in January.
The outtage comes at a particularly bad time for Salesforce, which had just issued a press release on Wednesday celebrating its near-flawless up-time in March.
Salesforce and other software as a service (SaaS) vendors were all the rage at Software 2006, a VC shindig that I attended this week out on Sandhill Road. SaaS deployments are a magic elixir that will solve costly headaches such as deployment, maintenance, upgrades and support. At the same time, SaaS gives ISVs a way around debilitating piracy.
But Salesforce and its cousins will have to be extremely attentive to service issues and interruptions going forward. While the Salesforce.com outtages don't seem to have a common cause, and may be expected from a company growing as fast as Salesforce is, knocking your entire North American customer base offline, for any reason at any time is untenable -- and won't engender a lot of good will.
Apologists for SaaS point out that traditional client/server software frequently crashes and experiences outtages as well. But companies such as Microsoft or SAP don't need to worry that all of their customers will be affected by an glitch at the same time.