timeouts are not reported as a failure_type |
|||||||
Issue descriptionBuild: https://ci.chromium.org/p/chromium/builders/luci.chromium.try/ios-simulator-full-configs/8739 buildbucket result: https://apis-explorer.appspot.com/apis-explorer/?base=https://cr-buildbucket.appspot.com/_ah/api#p/buildbucket/v1/buildbucket.get?id=8929159637564373040&_h=1& Expected result: results_detail_json should have a property that indicates the failure reason: \"failure_type\": \"TEST_TIMED_OUT\" example -- the following buildbucket has failure_type: TEST_FAILURE when the build fails due to a test failure: https://apis-explorer.appspot.com/apis-explorer/?base=https://cr-buildbucket.appspot.com/_ah/api#p/buildbucket/v1/buildbucket.get?id=8929376143997662912&_h=1&
,
Dec 18
,
Dec 19
@jbudorick can you ptal?
,
Dec 19
You've discovered the large difference between ios and non-ios infra. Does anything set TEST_TIMED_OUT? i.e., do non-ios tests do so? I'm not seeing it in https://codesearch.chromium.org/chromium/tools/depot_tools/recipes/recipe_modules/tryserver/api.py, which is where non-ios recipes set TEST_FAILURE.
,
Dec 19
or to rephrase: I don't think anything sets TEST_TIMED_OUT. Do we particularly care about having it? If so, I assume we want it for non-ios as well...
,
Dec 20
I think it's important to set a failure_type, since that can be used by monitoring tools to distinguish between different types of errors. In this case, swarming knows that the test suite timed out, but there's no way to know this by looking just at the buildbucket results.
,
Dec 20
Alright, this isn't iOS-specific, then, though the implementation for iOS will necessarily differ.
,
Dec 20
Updating the title to match #6; adding a new failure_type globally is a feature request.
,
Dec 21
|
|||||||
►
Sign in to add a comment |
|||||||
Comment 1 by erikc...@chromium.org
, Dec 4