Apex Code Development You need to sign in to do that
Don't have an account?
Kimberly Sullivan Failing Managed Package Unit Tests
Hello,
This has been a discussion many times at my office. We are a C# development shop mostly but about 2 years ago started supporting our salesforce org. We do a fair amount of development in apex now and have been doing lots of deployments as our organization is using Salesforce more and more. When we do a deployment into our full sandbox and into production org we always run all unit tests (not just local).
We have a couple managed packages in our org and once in a while there is an update and their unit tests fail. Our standard procedure when this happens is to halt all our deployments until we figure out how to fix the test or the support team for the app fixes the test. Does everyone operate this way? I've tried checking out SF best practices and don't see much on how many tests to run. Often we are told by app support to run local tests and continue with deployments but this seems risky. Other times we are told that our custom code is causing that problem. If our custom code can break a managed package unit test doesn't that mean our custom code could break the managed package?
How are other dev teams handling this?
Thank you in advance!
This has been a discussion many times at my office. We are a C# development shop mostly but about 2 years ago started supporting our salesforce org. We do a fair amount of development in apex now and have been doing lots of deployments as our organization is using Salesforce more and more. When we do a deployment into our full sandbox and into production org we always run all unit tests (not just local).
We have a couple managed packages in our org and once in a while there is an update and their unit tests fail. Our standard procedure when this happens is to halt all our deployments until we figure out how to fix the test or the support team for the app fixes the test. Does everyone operate this way? I've tried checking out SF best practices and don't see much on how many tests to run. Often we are told by app support to run local tests and continue with deployments but this seems risky. Other times we are told that our custom code is causing that problem. If our custom code can break a managed package unit test doesn't that mean our custom code could break the managed package?
How are other dev teams handling this?
Thank you in advance!