[SSSD] [PATCH] TESTS: extend PAM responder unit test

Lukas Slebodnik lslebodn at redhat.com
Thu Nov 5 07:59:32 UTC 2015


On (04/11/15 18:58), Pavel Reichl wrote:
>
>
>On 11/04/2015 06:41 PM, Sumit Bose wrote:
>>On Wed, Nov 04, 2015 at 06:00:34PM +0100, Pavel Reichl wrote:
>>>
>>>
>>>On 11/04/2015 03:02 PM, Lukas Slebodnik wrote:
>>>>On (04/11/15 14:40), Pavel Reichl wrote:
>>>
>>>>>
>>>>>Also one of few not yet tested corners of pam_reply() - 'Printing account
>>>>>expiration warning for sshd' requires pam_service to be sshd, but I can
>>>>>change when I will write another test (it might happen soon as pcech is
>>>>>working on similar change in adjacent code).
>>>>>
>>>>That's exactly the reason why the default value should not be sshd.
>>>
>>>I don't understand why this would be the reason can you elaborate?
>>
>>Because a default value should not lead to a code path which handles a
>>special case. The general PAM responder test should not run through the
>>'sshd' case in pam_reply() only if the service is set explicitly to
>>'sshd' this features should be tests.
>
>Sure, but just setting sshd would not lead to execution of this branch AFAIK
>at least pam verbosity would be needed to be set. As said before service
>value doesn't matter in current test so IMO this patch would only set service
>to more sane default however I agree with you that 'generic service name' is
>not a bad idea at all.
>
We spent enormous time with such unimportant patch.

I pushed patch with Sumit's propsal "pam_test_service"
using one-liner rule.

master:
* 8cfc3d5aeea21a3b8d1801aeef4866a4d3e7add0

LS


More information about the sssd-devel mailing list