Salesforce.com's reliable reputation as a software-as-a-service provider was tested Jan. 6 after thousands of business users were left without access to their enterprise applications for roughly 38 minutes, according to numerous reports on Twitter.
Salesforce.com, which hosts CRM and other enterprise applications over the Internet for thousands of companies, said on its Trust.salesforce.com community site that the problem thwarted over 177 million transactions in Europe, Asia and North America beginning around 20:39 GMT.
A core network device failed, stopping all data from being processed, according to the site, which provided details about the outage here. Salesforce.com staffers worked quickly to get the system up and running again and did so within an hour of the device failure.
"While we are confident the root cause has been addressed by the workaround, the Salesforce.com technology team will continue to work with hardware vendors to fully detail the root cause and identify if further patching or fixes will be needed," Salesforce.com said.
Such outages are not uncommon in the world of SAAS, or Internet-fueled computing, aka the cloud.
Google in the summer of 2008 suffered a few outages to Gmail and other Google Apps, prompting a loud outcry among thousands of users who use Google Apps to run their businesses. Google provided the affected Google Apps Premier Edition customers with credits.
But outages at Salesforce.com, the godfather of hosted CRM software, are less frequent. In fact, analysts have suggested Google should look to Salesforce.com as the model for cloud reliability.
After all, Salesforce.com has been hosting software for nearly a decade and it has largely proved a reliable alternative to on-premises solutions from Oracle, SAP, Microsoft and other enterprise application providers.
The outage will accomplish two things: give Oracle, SAP and Microsoft more firepower to argue that the cloud is still unreliable for enterprises, and force Salesforce.com, Google and other SAAS providers to redouble their efforts to get application service reliability as close to the 99.999 percent reliability benchmark as possible.