Dive Brief:

• Salesforce experienced a database failure on Tuesday that left some clients unable to access their services.
• The company said the issue was specific to one "instance" out of 45 of Salesforce’s instances, so only a small subset of customers were affected. 
• Salesforce said the failure was caused by a "file integrity issue" and was resolved by restoring the database from an earlier backup, though five hours of data could not be recovered and were permanently lost. 

Dive Insight:

The service had been down for at least 18 hours as of Wednesday morning. Eventually, the company said its database was back up, though it was still not operating at full capacity.

"The NA14 instance continues to operate in a degraded state," said a company update. "Customers can access the Salesforce service, but we have temporarily suspended some functionality such as weekly exports and sandbox copy functionality. In addition, search indexing may be delayed."

Salesforce customers complained about the outage on Twitter. Salesforce CEO Marc Benioff tweeted an apology for the issue, and the company’s website reported that their technology team has been working "around the clock" to fix it.

"At this point, we do not believe we will be able to repair the file integrity issue. To bring NA14 back to full health, we have shifted our focus to recovering from a prior backup, which was not impacted by the file integrity issues," according to the website.  

An article in The Register reported that the initial failure occurred at the Salesforce data center in Herndon, Va., and was followed by a failure at a Washington, D.C., facility.