What Went Down September 4th
Blog > Article

What Went Down: Week Ending September 4th

Check out “what went down” from Monday, August 29th to Sunday, September 4th.

 

Newsworthy: 

Microsoft Azure was fully down for some users starting around 11 AM Pacific time on August 30th and resolving around 9 AM on the 31st. The problem specifically affected the users that run Canonical Ubuntu 18.04. A “flawed security update” contributed to broken DNS queries.

Azure Umbutu Outage August 29

 

 

Notable Metrist-Reported Downtime

While these outages didn’t make the news, these issues caught by Metrist may have affected your company’s app and operations. 

 

AWS

  • AWS Cloudfront was down for 39 minutes in AWS-US-West-2 on 9/1, as the platform was unable to purge files for 24 minutes, then displayed extreme latency for 15 minutes.  

Azure

  • Azure App Service was unable to be reached in 5 of 12 North American regions for 8 hours and 49 minutes from 8/29 to 8/30. Additionally, the downtime could be possibly connected to DNS outage tied to an issue with Canonical Ubuntu 18.04. 
  • Azure DevOps Repos was down for 7 hours and 46 minutes on 8/29, corresponding with the Azure App Service outage. As a result, this downtime could also be tied to the Canonical Ubuntu 18.04 DNS outage. 

GCP

  • GCP Compute Engine was down for 15 hours and 20 minutes on 8/30. The app was unable to create new instances in GCP US East 4 due to a lack of capacity from GCP. 
  • Then on 8/31, it was down again for 17 hours and 20 minutes because it was unable to create new instances in GCP-US-East-4 due to a lack of capacity from GCP. 
  • GCP App Engine was down for 28 minutes on September 2nd, unable to create new versions from GCP-UW-West-2. 

 

GCP Cloud August 29th outage

Saas/Other

  • NuGet was down for 1 hr 18 mins on 8/31. The outage was characterized by a fluctuation between extreme latency and downtime when listing versions and downloading images from the AWS us-west-2 region.
  • Github was down for 13 minutes in the northeastern US, while it had extreme latency for all users in North America on 9/2.

Github Metrist Down August 29

 

Want to see more of “What Went Down”? Check out the archives here. 

And want to monitor cloud dependencies for yourself in less than a minute? Sign up for Metrist (it’s free!).

Subscribe to our newsletter

Follow along as we expand the definition of Observability

More articles

Need Help Locating an Article?

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