问题:

When running a service for a long time (a couple of days) I receive a connection timeout exception from Windows Azure Table Storage. Even the get table api returns this exception.

Why does it occur and how can I solve this?

回答1:

In addition to Gaurav's answer, the other common cause of this issue is port exhaustion. See http://blogs.msdn.com/b/windowsazurestorage/archive/2013/08/08/net-clients-encountering-port-exhaustion-after-installing-kb2750149-or-kb2805227.aspx for more information.

回答2:

One of the possible reason for this would be that your storage account is hitting the thresholds specified by Windows Azure Storage. Please ensure that you're not hitting those limits. Storage team has a blog about this which you can read here: http://blogs.msdn.com/b/windowsazurestorage/archive/2012/11/04/windows-azure-s-flat-network-storage-and-2012-scalability-targets.aspx. Another possibility could be that you are getting these as a result of transient errors.

First thing I would recommend is to enable storage analytics on your storage account, if not done already. That would give you a lot of information about what's going on at the storage level. You can read more about storage analytics here: http://msdn.microsoft.com/en-us/library/windowsazure/hh343270.aspx.

Second, in your code make sure that you have implemented a good retry policy. That would take care of transient errors.