Engineer’s typo cripples Amazon Web Services
Businesses choose cloud storage solutions because of the incredible uptime touted by the cloud providers. This week’s embarrassing, large scale outage for Amazon is a spectacular own goal.
An Amazon engineer was tasked with investigating a performance issue with their billing service but mis-typed part of a command and inadvertently took down more servers that intended.
This human error had knock on effects for a whole host of Amazon customers, and dented their reputation for highly available service provision.
Amazon say the duration of the outage was compounded because the indexing system used by AWS had not been rebooted for years. A lesson learnt there then.
It’s a salient reminder that the human element is often the weakest link and that no matter what the marketing materials say, cloud storage is not fallible.