Answered How do I know if I'm being throttled on storage?

  • Tuesday, March 27, 2012 1:19 PM
     
     
    In the past few days I've started to routinely receive connection refused errors when trying to access Azure storage (queues more often than anything).  Originally it was occurring in my node.js code I thought it was just something I had messed up in my code or it was a problem in my environment.  Unfortunately the problem has persisted and I've experienced it at multiple locations and machines (the code in question does not run in an Azure compute instance, it instead runs on remote boxes outside of Azure) and I've experienced the same problem from a third party tool (Cerebrata Cloud Storage Studio when trying to access the queues).  The failures are intermittent, but are happening with enough frequency that it is quite painful.  I'm at a loss as to why this is occurring and started to theorize this morning that perhaps I'm being throttled due to my activity levels (I've done  1070 batches of 10,000 transactions within the last ten days).  Does that number seem high enough to throttle?  If so is there anyway I can tell whether my account is being throttled?

All Replies

  • Tuesday, March 27, 2012 1:23 PM
     
     Answered

    Stated SLA is 5000 storage transactions per account per second. 10M transactions in 10 days is very reasonable for Azure Storage to handle, unless you're trying to process them all in 10 minutes or some other short amount of time.

    Chances are that you're not being throttled, but Azure Storage is having intermittent issues or your code has issues.  I'd suggest contacting support.


    Auto-scaling & monitoring service for Windows Azure applications at http://www.paraleap.com



  • Tuesday, March 27, 2012 2:20 PM
     
     

    Are you using a storage account in North Central?  If so, you are likely in same boat as everyone else in that sub-region.  It has been extraordinarily slow for almost 24 hrs now.  Nothing on dashboard yet, but suspect they are having a 'service degradation'.


    Ryan Dunn --  http://opstera.com


    • Edited by dunnry Tuesday, March 27, 2012 2:20 PM
    •  
  • Tuesday, March 27, 2012 3:01 PM
     
     

    Hmm, no, I'm south central, I've been checking the health dashboard, but nothing has been showing...

  • Tuesday, March 27, 2012 10:51 PM
     
     

    Are you using a storage account in North Central?  If so, you are likely in same boat as everyone else in that sub-region.  It has been extraordinarily slow for almost 24 hrs now.  Nothing on dashboard yet, but suspect they are having a 'service degradation'.


    Ryan Dunn --  http://opstera.com


    Agreed.  I'm seeing timeouts on top of timeouts and support has come back with a statement that the frequency of errors is within SLA :|


    Auto-scaling & monitoring service for Windows Azure applications at http://www.paraleap.com

  • Wednesday, March 28, 2012 11:17 AM
    Moderator
     
     Answered

    Hi,

    If you think there’re some issues with the storage service, please try to find the request ID in a problematic request, and then contact our customer support on https://support.microsoft.com/oas/default.aspx?gprid=14928&st=1&wfxredirect=1&sd=gn. Many classes in the storage client library contain a GetRequestId method.

    Best Regards,

    Ming Xu.


    Please mark the replies as answers if they help or unmark if not.
    If you have any feedback about my replies, please contact msdnmg@microsoft.com.
    Microsoft One Code Framework