Due to network difficulties, Gandi's services are temporarily unavailable for some of you. Please accept our apologies for any inconvenience this may have caused you. We are currently working on resolving the problem as quickly as possible.

UPDATE 17.30 GMT: The incident has now been resolved. We have fixed the equipment responsible for the network access problems.

Here is a clear and open explanation of the event:

"General" version:

Due to a configuration error on our part (human error), we started a chain reaction with a serious routing bug on our network equipment. This led to the system being ground to a halt, which is why our services were unavailable for one quarter of the service providers.

"Expert version":

The involuntary passage of Cisco's switching software led to half of the CEF routes being rendered unavailable, despite an immediate return to the original configuration.

We offer our most sincere apologies for this incident, which does not resemble us at all. Of course, we will grant a refund of one hosting day for customers that feel wronged (which is the reason this will be granted upon request by the way), and will take all the necessary actions to assure that this incident will just remain a bad memory.