MainThreadMonitor: don't crash if there is no test in progress
#2838
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
NachoSoto
force-pushed
the
main-thread-monitor-during-tests
branch
from
dfc1d64
to
1ec112b
Compare
I was testing this locally, and I noticed that sometimes this would trigger _after_ a test had already finished: ``` Test Case '-[BackendIntegrationTests.StoreKit2IntegrationTests testCanPurchasePackage]' passed (16.712 seconds). Test Suite 'StoreKit2IntegrationTests' passed at 2023-07-18 10:27:45.282. Executed 1 test, with 0 failures (0 unexpected) in 16.712 (16.713) seconds Test Suite 'BackendIntegrationTests.xctest' passed at 2023-07-18 10:27:45.282. Executed 1 test, with 0 failures (0 unexpected) in 16.712 (16.713) seconds Test Suite 'Selected tests' passed at 2023-07-18 10:27:45.282. Executed 1 test, with 0 failures (0 unexpected) in 16.712 (16.714) seconds BackendIntegrationTests/MainThreadMonitor.swift:47: Precondition failed: Main thread was blocked for more than 1.0 seconds 2023-07-18 10:27:46.301255-0700 BackendIntegrationTestsHostApp[28299:2377529] BackendIntegrationTests/MainThreadMonitor.swift:47: Precondition failed: Main thread was blocked for more than 1.0 seconds ``` Looking at what the main thread was doing, I noticed `XCTest` was blocked waiting for some internal signal, possibly at the end of a test suite. We obviously don't want to consider that a deadlock, since it's not our fault, so I changed the implementation to ignore this if the test isn't in progress.
NachoSoto
force-pushed
the
main-thread-monitor-during-tests
branch
from
1ec112b
to
1430994
Compare
Codecov Report
@@ Coverage Diff @@
## main #2838 +/- ##
==========================================
+ Coverage 86.57% 86.58% +0.01%
==========================================
Files 217 217
Lines 15513 15513
==========================================
+ Hits 13430 13432 +2
+ Misses 2083 2081 -2 |
aboedo
approved these changes
Jul 18, 2023
NachoSoto
added a commit
that referenced
this pull request
Jul 19, 2023
**This is an automatic release.** ### Dependency Updates * Bump fastlane from 2.213.0 to 2.214.0 (#2824) via dependabot[bot] (@dependabot[bot]) ### Other Changes * `MainThreadMonitor`: don't crash if there is no test in progress (#2838) via NachoSoto (@NachoSoto) * `CI`: fixed Fastlane APITester lanes (#2836) via NachoSoto (@NachoSoto) * `Integration Tests`: workaround Swift runtime crash (#2826) via NachoSoto (@NachoSoto) * `@EnsureNonEmptyArrayDecodable` (#2831) via NachoSoto (@NachoSoto) * `iOS 17`: added tests for simulating cancellations (#2597) via NachoSoto (@NachoSoto) * `CI`: make all `Codecov` jobs `informational` (#2828) via NachoSoto (@NachoSoto) * `MainThreadMonitor`: check deadlocks only ever N seconds (#2820) via NachoSoto (@NachoSoto) * New `@NonEmptyStringDecodable` (#2819) via NachoSoto (@NachoSoto) * `MockDeviceCache`: avoid using real `UserDefaults` (#2814) via NachoSoto (@NachoSoto) * `throwAssertion`: fixed Xcode 15 compilation (#2813) via NachoSoto (@NachoSoto) * `CustomEntitlementsComputation`: fixed API testers (#2815) via NachoSoto (@NachoSoto) * `PackageTypeTests`: fixed iOS 12 (#2807) via NachoSoto (@NachoSoto) * `Tests`: avoid race-condition in leak detection (#2806) via NachoSoto (@NachoSoto) * Revert "`Unit Tests`: removed leak detection" (#2805) via NachoSoto (@NachoSoto) * `PackageType: Codable` implementation (#2797) via NachoSoto (@NachoSoto) * `SystemInfo.init` no longer `throws` (#2803) via NachoSoto (@NachoSoto) * `Trusted Entitlements`: add support for signing `POST` body (#2753) via NachoSoto (@NachoSoto) * `Tests`: unified default timeouts (#2801) via NachoSoto (@NachoSoto) * `Tests`: removed forced-unwrap (#2799) via NachoSoto (@NachoSoto) * `Tests`: added missing `super.setUp()` (#2804) via NachoSoto (@NachoSoto) * Replaced `FatalErrorUtil` with `Nimble` (#2802) via NachoSoto (@NachoSoto) * `Tests`: fixed another flaky test (#2795) via NachoSoto (@NachoSoto) * `TimingUtil`: improved tests by using `Clock` (#2794) via NachoSoto (@NachoSoto) * `IgnoreDecodeErrors`: log decoding error (#2778) via NachoSoto (@NachoSoto) * `TestLogHandler`: changed all tests to explicitly deinitialize it (#2784) via NachoSoto (@NachoSoto) * `LocalReceiptParserStoreKitTests`: fixed flaky test failure (#2785) via NachoSoto (@NachoSoto) * `Unit Tests`: removed leak detection (#2792) via NachoSoto (@NachoSoto) * `Tests`: fixed another flaky failure with asynchronous check (#2786) via NachoSoto (@NachoSoto) --------- Co-authored-by: NachoSoto <ignaciosoto90@gmail.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
I was testing this locally, and I noticed that sometimes this would trigger after a test had already finished:
Looking at what the main thread was doing, I noticed
XCTestwas blocked waiting for some internal signal, possibly at the end of a test suite.We obviously don't want to consider that a deadlock, since it's not our fault, so I changed the implementation to ignore this if the test isn't in progress.
I'd bet this is the last time (🤞🏻 ) I have to change this. This follows up several PRs trying to avoid false positives in CI:
MainThreadMonitor: check deadlocks only ever N seconds #2820MainThreadMonitor: increased threshold #2662MainThreadMonitor: fixed flakiness in CI #2517