Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.Sign up
Make RoundhousE fault tolerant to transient exceptions that often occur for Azure Database #234
We're using RoundhousE to manage migrations on Azure Database. As it occurred, transient faults (exceptions that could be easily handled by retrying the operation) are often happen in cloud environment.
Here the typical places where exceptions occur:
It would be great if RoundhousE had a built-in ability to handle such exceptions via some retry policy.
Below are some of the exceptions we've seen so far:
Timeout on opening connection:
Network issue when executing an update script:
Internal server error when creating a database:
Semaphore timeout when executing an update script:
Exception when retrieving list of already executed scripts:
As far as I understand there is only simple retry for just a single case - https://github.com/chucknorris/roundhouse/blob/master/product/roundhouse/databases/AdoNetDatabase.cs#L139