Jest step sometimes doesn't run tests

#1

Jest step sometimes doesn’t run tests

Description of the issue

Build sometimes fails as it attempts to deploy the lcov_report generated when running jest but cannot find the file. Turns out the tests aren’t running at all to actually generate the file.

The testing step runs our yarn test:ci script which runs jest --debug --verbose --ci It starts off outputting the usual log stuff then just ends there without running a single test.

Oddly, this seems to be random and temperamental, and simply running a rebuild solves it for that build as it then runs the tests properly and passes fine. We can then go several builds and days before it happens again.

It doesn’t seem to matter what branch, how big the PR is, date & time, etc. Just seems to be random.

The steps used are fairly standard:

  • activate-ssh-key
  • git clone repo
  • continue from repo
  • cache-pull
  • yarn install
  • create .env files
  • yarn lint (runs eslint .)
  • yarn typecheck (runs tsc -p tsconfig.spec.json)
  • Run tests - this is a script that runs the following
    • set -o pipefail
    • yarn test:ci (runs jest --debug --verbose --ci) <-- issue, doesn’t run any tests but outputs the pre-test log followed by Done in X seconds
    • tee simple-test-report.txt
  • Deploy the simple-test-report
  • Deploy the full lcov_report <-- fails due to no lcov_report generated
  • cache-push
  • Message Slack

Reproducibility

  • Does a “Rebuild” help? (You can trigger a rebuild from the Build’s page, by clicking the “Rebuild” button in the top right corner of a finished build) : YES
  • Does a rebuild without caches help? (You can remove the Cache:Pull and Cache:Push steps temporarily to not to use the cache, or you can delete all the caches on the Settings tab of the app. : YES but probably just because of the rebuild
  • Does the issue happen sporadically, or every time? : SPORADICALLY
  • When did the issue start? : Around 16th June 2020

Local reproduction

Running the yarn test:ci jest script locally works as expected, tests are run and the coverage report is generated. Running that specific test script locally, mentioned above with the test report file, works fine too. As do all the other steps that can be run locally.

Running the bitrise pipeline locally via CLI also seems to work fine, though due to the random nature of this issue it might require many local runs before it appears.

0 Likes

closed #2

This topic was automatically closed 30 days after the last reply. New replies are no longer allowed.

0 Likes