AWS on Friday suffered a routing failure that came across as a little déjà vu, the day after Microsoft Azure also cut customers off from a major cloud region even though workloads continued to run.

AWS blamed networking hardware affecting traffic into and out of us-west-2, while Microsoft earlier said an automation bug removed more Azure West US routes than intended during maintenance.

See also: A network maintenance error separated an Azure West data center from the internet for hours

The AWS failure began at 10:55 UTC on 24 July and affected internet connectivity, Direct Connect and several services in us-west-2. AWS traced it to networking hardware serving traffic into and out of the region, with customers using Direct Connect through the EqSe2 exchange in Seattle experiencing an extended disruption. A mitigation restored most connectivity, although route reconvergence caused a second drop between 11:47 and 11:59 UTC.

A day earlier, Microsoft had isolated an Azure West US datacentre from its wider network during routine fibre maintenance. It said a bug in its automated request-conversion system removed IP routes from more devices than intended, disrupting ingress and egress. Microsoft began rolling back the change at 17:45 UTC and restored the network by 18:26 UTC.

Redundancy fails

Get the full story: Subscribe for free

Join peers managing over $100 billion in annual IT spend and subscribe to unlock full access to The Stack’s analysis and events.

Subscribe now

Already a member? Sign in