2

Test execution stops after 2-3 hours on iOS 11 devices. The same setup works fine for iOS 10 devices. gems being used - run_loop - 2.6.1/2.6.0/2.5.4 calabash-cucumber - 0.21.1/0.20.5 cucumber - 3.0.0.pre.2 Xcode - 9 iOS - 11.0/11.0.1/11.0.2

Below is the error message -

Could not install /Users/raenug001c/.rvm/gems/ruby-2.3.1@ios11test/gems/run_loop-2.6.1/lib/run_loop/device_agent/ipa/DeviceAgent-Runner.app. iOSDeviceManager says:

 Error getting device with ID 630826b68345ba162bc2494709ff39657dd3c122: Error Domain=com.facebook.FBControlCore Code=0 “Failed to load device console entries” UserInfo={NSLocalizedDescription=Failed to load device console entries}

   install <app-path>
            -d,--device-id  <device-identifier>     iOS Simulator GUID or 40-digit physical device ID
            -u,--update-app <should-update-app> [OPTIONAL]  When true, will reinstall the app if the device contains an older version than the bundle specified     DEFAULT=1
            -p,--profile-path       <profile-path> [OPTIONAL]       Path to provisioning profile
            -c,--codesign-identity  <codesign-identifier> [OPTIONAL]        Identity used to codesign app bundle [device only]. Deprecated - should use profile path.       DEFAULT=
            -i,--resources-path     <resources-path> [OPTIONAL]     Path to resources (executables) to inject into app directory. A list of colon separated files may be specified.

  (RuntimeError)
  /Users/raenug001c/Documents/testios11/tvx-mobile-automation/iOS/common/support/01_launch.rb:49:in `Before’

Only after the restarting the device, the tests can be executed further.

After further debugging the issue, seems like iOSDeviceManager is unable to install/launch device agent app on the device. Unable to view the source code for iOSDeviceManager, hence not able to debug further.

Ram
  • 65
  • 4
  • 12
  • We are aware of the problem, but don't have solution. We think this is probably an Apple bug. Put another way, if you were to run an XCUITest for more than 2 hours, you would see the same result. We are tracking this problem internally. – jmoody Oct 24 '17 at 06:58

0 Answers0