MICROSOFT'S OUTLOOK.COM went kaput for a few hours last night, but there was plenty of confusion before Microsoft updated its Office 365 service site and admitted to the fault.
The Outlook.com website started throwing up error messages from around 5pm GMT/10am Pacific Time.
The outage lasted about three hours on and off, and Microsoft's Office 365 service health site was initially slow to update, announcing that 'Everything is up and running' even though there was clearly a problem.
Meanwhile, Twitter started to fill up with hundreds of messages complaining about the outage.
#ALERT --> Windows Live Hotmail is down or having service trouble !!— Ricardo Wagner © (@ricardowec) October 6, 2016
Microsoft eventually admitted that "users may be unable to connect to the Outlook.com service" and that the firm was "analysing the affected infrastructure to determine the cause".
This turned out to be a portion of the capacity responsible for routing connection requests to the mailbox databases becoming "degraded", which Microsoft fixed by shifting traffic to a "healthy portion of the infrastructure to remediate impact".
Microsoft announced at 8pm UK time that Outlook.com was back up and running.
The outage didn't affect everybody and was probably dependent on which of those "degraded" connections a customer's box was attached to. It was mainly confined to the US, but at least one member of INQ staff had access troubles last night.
Microsoft service outages are never a light matter, as growing reliance on cloud ecosystems such as Office 365 mean a greater risk to enterprise productivity.
As Mimecast commented to us last time there was a major Azure outage, things can really only get worse in this regard. µ
A surprisingly busy week in a quiet month
Measures just 15.75mm at its thickest point
Firm expects GPU sales to start drying up