Blog Post

Azure Lab Services Blog
1 MIN READ

Azure Lab Services - performance issues in UK South region - January 21, 2025

Stan_Ageev's avatar
Stan_Ageev
Icon for Microsoft rankMicrosoft
Jan 21, 2025

We are currently experiencing performance issues in the UK South region.  That may cause some operations to take longer than usual to complete. This delay can result in certain virtual machines becoming stuck in the starting or stopping state.

 

Our engineering team is actively working on mitigating the issue and has started rolling out several changes aimed at improving performance. We are also closely monitoring the region and unblocking affected virtual machines as soon as any issues are detected.

 

For new labs, we recommend selecting a different region (other than UK South) until the issue is fully resolved. We will provide the next status update by January 24, 2025.

 

We sincerely apologize for any inconvenience this may cause and deeply appreciate your understanding and patience as we work to restore optimal performance. Thank you for choosing our services.

Published Jan 21, 2025
Version 1.0

2 Comments

  • We have identified the issue to be a code bug in a dependency, which resulted in stuck executions for workflows in the region. The issue in the dependency was fixed and deployed on January 28, 2025 at 1 PM PT. Since the deployment, we have observed recovery in the region and will continue to monitor for any regressions.

  • We have identified a set of virtual machines with long event histories impacting performance in the region. A mitigating action has been taken to delete the event histories for these virtual machines. Though we’re still gathering data, preliminary indications show that performance has significantly improved as a result.

     

    We are committed to ensuring that all affected virtual machines are unblocked as soon as any issues are detected. We will provide the next status update by January 28, 2025.