Final Update: Thursday, September 14th 2017 13:10 UTC
VSTS in the South Central US region is now back in a healthy state. The incident was mitigated after the bad web tier node was taken out of the load balancer. We've confirmed that users are no longer seeing slowness or errors. We are working to analyze the memory dump from the bad web node to understand root cause, so we can be sure the issue doesn't happen again.
Update: Thursday, September 14th 2017 10:45 UTC
DevOps team continues to investigate issues with VSTS services across South Central US region. Our logs indicate that the issue started on September 14th, 2017 at 09:10 UTC. Initial investigation shows that the issue is because of a problematic web role and we have removed that instance from the load balancer. Team is monitoring the system while we continue to analyze the captured dumps and traces to find the root cause.
- Next Update: Before Thursday, September 14th 2017 13:00 UTC
Initial Update: Thursday, September 14th 2017 09:45 UTC
We are investigating performance issues affecting users of the service. Users in SCUS may experience degraded performance or slow page load times.
- Next Update: Before Thursday, September 14th 2017 10:45 UTC