-
Notifications
You must be signed in to change notification settings - Fork 113
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Improve test run time #92
Comments
* Transform cypress tests (opensearch-project#21) * Added Cypress tests for Transforms UI * License update * Unit tests initial commit * Added jest tests for Transforms, details, edit and create * Sample transform fix * Fixed cypress test transform and sample data load order Signed-off-by: Eric Lobdell <[email protected]> * Test fix debug * Simplified unit test * Updated transforms cypress test comments and success criteria Signed-off-by: Eric Lobdell <[email protected]> * Fixed quotation mark issue in test Signed-off-by: Eric Lobdell <[email protected]> * Removed debug mapping and index objects from unit test Signed-off-by: Eric Lobdell <[email protected]> * Cypress test improvements and completed edit transform unit test Signed-off-by: Eric Lobdell <[email protected]> * Integ test delete fix and EditTransforms jest test fix Signed-off-by: Eric Lobdell <[email protected]> * Restoring enable disable page refresh delay to fix failing integ test Signed-off-by: Eric Lobdell <[email protected]> * Resolve test constants conflict Signed-off-by: Eric Lobdell <[email protected]>
So the majority of the long test times comes from the fact that under the hood we are using the testing framework from core that uses the randomization runner which means we can't parallelize our tests (even though the majority if not all could be parallelized). So as Ravi pointed out we should look into splitting the tests by feature first and then from there we could further split each feature into a subset of tests (such as different REST APIs and other things in ISM). |
Is your feature request related to a problem? Please describe.
At the moment running the tests of IndexManagement takes a long time. Need to bring this time down to make the code development process easier
Describe the solution you'd like
Make the different test suites run as different github actions so PRs don't need to wait around 20-30 min to see if the tests succeed. We can parallelize the actions at the current highest feature level.
Test cleanup - any test that can be an unit test should be one, go through existing tests and move any integration tests that can be just unit tests
The text was updated successfully, but these errors were encountered: