Down outage what went down image
Blog > Article

What Went Down: Cloud App Outages from Metrist

What went down this week? There were a number of newsworthy and noteworthy outages for the cloud apps our platforms depend on. 

Newsworthy: 

Some of the outages to make the headlines this week were from heavy hitters GCP, Oracle, and Microsoft 365. 

  • Heatwave leads to GCP and Oracle Outages. Google Cloud and Oracle chose to shut down equipment in some of their European data centers due to cooling apparatuses failing during the prolonged heat wave on Tuesday, July 19th. The companies chose to shut down zone europe-west2-a for region europe-west2 (GCP) and UK Data Center (Oracle) temporarily until temperatures could be stabilized. They decided to take this step in lieu of allowing equipment to overheat and become permanently damaged. Google restored its services on Tuesday at 11:45 PM EST and Oracle was back up by Wednesday at 7:00 AM EST.
  • Multiple Microsoft 365 services down worldwide. A flawed Enterprise Configuration Service (ECS) deployment led to Microsoft 365 outage across the world on Thursday, July 21st. The 5-hour downtime started with Microsoft Teams and eventually affected Exchange, Word, Forms, Graph API, Office, SharePoint, PowerPlatform, Autopatches, and Yammer to varying degrees. The issues started at approximately 1:05 AM UTC and most issues were resolved by 6:00 AM UTC. 

 

Notable: 

These outages didn’t make the news, but they may have affected your company’s app and operations. 

  • GCP App Engine was degraded for over an hour on July 20th due to extreme latency in the us-central1.
  • AWS CloudFront experienced off and on latency spikes all week for PurgeFile requests. Users in CA-Central-1 were also unable to PurgeFiles at all for about 5 minutes on July 20th from 3:25-3:30 PM UTC.
  • Azure SQL was unable to create new SQL Servers in the East US region on the 19th for 7 minutes. The app also experienced a number of issues over the course of the week, including frequent slowness from the 20th – 22nd. 
  • Azure Active Directory was partially down in US West 2 from 6:25-6:31 PM UTC on July 21st. 
  • GCP Compute Engine was partially down for nearly an hour (4:02-4:43 UTC) on July 23rd when users were unable to delete instances in North America Northeast 2 
  • Bambora had momentary outages that lasted approximately 3 to 8 minutes on the 23rd and the 25th of July.
  • CircleCI was partially down for about an hour and a half on July 19th and about a half hour on the 20th. (See below for the Metrist notification.)

 

Apps go down, but as long as we’re aware and have a backup plan, our companies can be more resilient. Have questions about this list? Contact us. 

If you’d like to keep track of the apps you depend on in real-time, try Metrist.

Subscribe to our newsletter

Follow along as we expand the definition of Observability

More articles

What Went Down Metrist August 15
What Went Down
Ellen Steinke

What Went Down August 15, 2022

This week there were a number of newsworthy and notable issues with the cloud apps our companies depend on. Check out “what went down” from

Read More »

Need Help Locating an Article?

Try using our search feature to find the post you're looking for.