AllocationFailed -- Pool unable to allocate nodes using built-in image RRS feed

  • Question

  • I have a compute pool on Azure Batch that is unable to allocate nodes. The error code is "AllocationFailed" and the message is "Desired number of dedicated nodes could not be allocated as they are not available." Resizing via Batch Explorer causes it to momentarily try to resize, but then it goes back to "AllocationFailed Desired number of dedicated nodes could not be allocated as they are not available." I'm using a built-in VM image (Ubuntu LTS-18 by Canonical). It's for dedicated cores, and the batch service is well under the quota for dedicated cores. I've had numerous other compute pools on the same batch service work fine (dozens in the past, half a dozen at the same time as this one failed).

    Any ideas on what might be causing this behavior and how to prevent it in the future? Thanks.

    Thursday, June 6, 2019 6:27 PM

All replies

  • Generally this errors occurs if Azure is unable to allocate the correct number of nodes you are requesting in a specific data center. It usually can be fixed by simply trying the create option again. 

    Another thing to try would be another size/ region. 

    I would be curious to see if the same create option works using another size/ region. 

    Thursday, June 6, 2019 7:17 PM
  • Any update on this? 

    Friday, June 14, 2019 10:38 PM
  • Is there any update on this issue? If the suggested answer has solved your issue, do click on "Mark as Answer" and “Vote as Helpful” on the post that helps you, this can be beneficial to other community members.

    Karishma Tiwari - MSFT

    Tuesday, July 2, 2019 8:54 PM