[Bug]: Chrome is automatically getting closed in the middle if any testcase got failed

[Kartheeswari-git]

Version

1.49.0

Steps to reproduce

Issue: Lets say I have 10 testcases. During the execution, for some reason testcase#5 failed and now its closing the crome and opening the new one to execute the testcase#6.

Expected behavior

It should execute testcase #6 from the same chrome where testcase#5 left instead of closing it. Its working like this in Playwright python and facing the issue with Playwright Js

Actual behavior

Its closing the chrome if any testcase got failed which is affecting all the upcoming testcases in the suite.

Additional context

No response

Environment

System:
    OS: Windows 10 10.0.19045
    CPU: (8) x64 11th Gen Intel(R) Core(TM) i5-11300H @ 3.10GHz
    Memory: 5.90 GB / 19.74 GB
  Binaries:
    Node: 22.11.0 - C:\Program Files\nodejs\node.EXE
    Yarn: 1.22.19 - ~\AppData\Roaming\npm\yarn.CMD
    npm: 10.9.1 - ~\node_modules\.bin\npm.CMD
  IDEs:
    VSCode: 1.95.3 - C:\Users\Karthee\AppData\Local\Programs\Microsoft VS Code\bin\code.CMD
  Languages:
    Bash: 5.1.4 - C:\Windows\system32\bash.EXE
  npmPackages:
    @playwright/test: ^1.49.0 => 1.49.0
    playwright: ^1.49.0 => 1.49.0

[Skn0tt]

Hi there! Playwright's working as intended here. Could you elaborate on why it's important to your usecase that chrome's not restarted?

[Kartheeswari-git]

I want to avoid restarting Chrome on failure and instead maintain the same Chrome context throughout the test suite execution. Here's the scenario:

We execute our product test cases in the following structure:

Suite1: 56 test cases
Suite 2: 30 test cases

Each suite contains test cases for specific pages, combining both UI validation and functionality tests. During execution, if an issue occurs (e.g., the UI doesn't display the expected error message), a particular test case might fail. However, the subsequent test case is supposed to continue from the same page where the previous one left off.

Currently, because Chrome is being closed and reopened upon failure, the test loses the page state required for the next execution. Setting this in a "Before All" method isn't feasible since each test case starts on a different state.

[AnzeKop]

Running into a similar issue in our app, we want to have 1 persistant browser assigned to the BrowserManager class but as the app can hit errors we don't want this to restart the browser as it leads to this error when being used

Browser logs:

<launching> /Users/anze/Library/Caches/ms-playwright/chromium_headless_shell-1148/chrome-mac/headless_shell --disable-field-trial-config --disable-background-networking --disable-background-timer-throttling --disable-backgrounding-occluded-windows --disable-back-forward-cache --disable-breakpad --disable-client-side-phishing-detection --disable-component-extensions-with-background-pages --disable-component-update --no-default-browser-check --disable-default-apps --disable-dev-shm-usage --disable-extensions --disable-features=ImprovedCookieControls,LazyFrameLoading,GlobalMediaControls,DestroyProfileOnBrowserClose,MediaRouter,DialMediaRouteProvider,AcceptCHFrame,AutoExpandDetailsElement,CertificateTransparencyComponentUpdater,AvoidUnnecessaryBeforeUnloadCheckSync,Translate,HttpsUpgrades,PaintHolding,ThirdPartyStoragePartitioning,LensOverlay,PlzDedicatedWorker --allow-pre-commit-input --disable-hang-monitor --disable-ipc-flooding-protection --disable-popup-blocking --disable-prompt-on-repost --disable-renderer-backgrounding --force-color-profile=srgb --metrics-recording-only --no-first-run --enable-automation --password-store=basic --use-mock-keychain --no-service-autorun --export-tagged-pdf --disable-search-engine-choice-screen --unsafely-disable-devtools-self-xss-warnings --enable-use-zoom-for-dsf=false --use-angle --headless --hide-scrollbars --mute-audio --blink-settings=primaryHoverType=2,availableHoverTypes=2,primaryPointerType=4,availablePointerTypes=4 --no-sandbox --disable-gpu --disable-dev-shm-usage --disable-software-rasterizer --disable-extensions --no-sandbox --single-process --user-data-dir=/var/folders/zb/hgzc5jv96xg8sjcm44y344yw0000gn/T/playwright_chromiumdev_profile-W8Cz0p9tkW8s --remote-debugging-pipe --no-startup-window
<launched> pid=13433
[pid=13433][err] [1204/140344.549627:WARNING:viz_main_impl.cc(85)] VizNullHypothesis is disabled (not a warning)
[pid=13433][err] [1204/140344.575595:ERROR:cv_display_link_mac.mm(165)] CVDisplayLinkCreateWithCGDisplay failed. CVReturn: -6670

[Skn0tt]

Tests in Playwright are meant to be executed independently from one another. If one of your tests depends on the state from a previous one, that sounds like it should be one combined test instead.

During execution, if an issue occurs (e.g., the UI doesn't display the expected error message), a particular test case might fail. However, the subsequent test case is supposed to continue from the same page where the previous one left off.

I think you're looking for soft assertions: https://playwright.dev/docs/test-assertions#soft-assertions

Running into a similar issue in our app,

This looks like a separate issue. If you believe there's a bug in Playwright here, please open a separate issue.

Restarting the browser after a test failure is intended, so everything's working as expected here. Closing.