Amazon said that about 0.07 percent of the data storage in its restored eastern cloud service isn’t fully recoverable. The company’s cloud-based web services suffered an outage that lasted for three-and-a-half days last week and service was restored for the most part on Monday.

But some of the data — 0.07 percent of the eastern region — hasn’t been recovered, and the company is contacting the limited number of customers who have lost data. One of the lessons being learned here is that backup sometimes isn’t enough, and customers who use web services may have to use multiple cloud vendors in the future.

The Amazon crash has raised questions about how much users will be able to trust cloud storage. Some companies using Amazon’s service are offering their own customers freebees to make the hassle a bit more palatable. For example, HootSuite is offering $50 credit for customers affected by the Amazon downtime.

The outage, which happened at the same time Sony suffered a hacker attack that brought down its PlayStation Network, brought down numerous high-profile Amazon customers, including Quora, Sencha, Reddit and Foursquare. Amazon said that a detailed post mortem is coming for the Elastic Block Store outage, and it is looking into the root causes. One status message report said that a “network event” caused the service to “re-mirror” a large number of EBS data in the eastern region.

On Monday, Amazon reported that its profits fell 33 percent as it spent to build more more and more warehouses.