Repository navigation
Set-Service.Tests.ps1 tests Failing builds #4800
Description
#4785 Introduced tests which are failing builds.
On Windows:
Description: New-Service with bad parameters will fail for 'badstarttype' where 'StartupType' = 'System'
Name: Set/New-Service cmdlet tests..New-Service with bad parameters will fail for 'badstarttype' where 'StartupType' = 'System'
message:
Expected string length 66 but was 13. Strings differ at index 0.
Expected: {CouldNotNewService,Microsoft.PowerShell.Commands.NewServiceCommand}
But was: {No Exception!}
-----------^
stack-trace:
at line: 65 in C:\projects\powershell\test\tools\Modules\HelpersCommon\HelpersCommon.psm1
65: $_.FullyQualifiedErrorId | Should Be $FullyQualifiedErrorId | Out-Null
On Linux and macOS:
[-] Error occurred in Describe block 79ms
The term 'Get-Service' is not recognized as the name of a cmdlet, function, script file, or operable program. Check the spelling of the name, or if a path was included, verify that the path is correct and try again.
At /home/travis/build/PowerShell/PowerShell/test/powershell/Modules/Microsoft.PowerShell.Management/Set-Service.Tests.ps1:25 char:48
System is a valid System.ServiceProcess.ServiceStartMode so either the test is flawed or there is an issue with New-Service.
The Linux/macOS failures are due to Get-Service being called in the TestCases HashTable literal. That literal gets parsed regardless of the OS skip resulting in calls to Get-Service on OSE's where it is not present.
Activity
iSazonov commentedon Sep 10, 2017
Mark Kraus (@markekraus) Thanks for your report - we skipped
[Feature]in last commits 😕markekraus commentedon Sep 10, 2017
Ilya (@iSazonov) There are some other PRs "in the pipe" that look like they affect feature tests but don't have any CI runs with
[feature]Also, feature test fails on Linux and macOS continue to show "passed" re: #4641 (example).☹️
iSazonov commentedon Sep 10, 2017
I see that "feature test fails on Linux and macOS continue to show "passed" "
/cc Steve Lee (@SteveL-MSFT) Travis Plunk (@TravisEz13)
TravisEz13 commentedon Sep 11, 2017
The code has been changed so that only non-PRs will fail due to Pester test failures. I didn't notice that logic error last time.
markekraus commentedon Sep 11, 2017
Travis Plunk (@TravisEz13) Now that #4807 is merged... I predict the nightly will fail for Linux and macOS due to This little nugget. example