Another user at realm type issue/bug

[Noel Power]

On 30/09/16 15:07, Andrew Bartlett wrote:
> On Thu, 2016-09-29 at 23:14 +0200, Andreas Schneider wrote:
>> On Thursday, 29 September 2016 12:28:20 CEST Jeremy Allison wrote:
>>> On Thu, Sep 29, 2016 at 07:22:28PM +0100, Noel Power wrote:
>>>> On 29/09/16 18:54, Jeremy Allison wrote:
>>>>> On Thu, Sep 29, 2016 at 06:10:55PM +0100, Noel Power wrote:
>>>>>> Hi,
>>>>>>
>>>>>> When looking at the parse_domain_user issue with user at realm
>>>>>> credentials
>>>>>> I was playing around with smbtorture and noticed that the
>>>>>> cli_credentials.username field is not setup when using -Uuser
>>>>>> @realm
>>>>>> with
>>>>>> smbtorture, this seems like a bug to me, please see that
>>>>>> attached patch
>>>>> Shouldn't this just be removing the "return;" ?
>>>>>
>>>>> The 'goto done' and new done: label is redundent here.
>>>> doh! /me face palms
>>>> you are completely correct :-)
>>>>
>>>> here's v2
>>> Reviewed-by: Jeremy Allison <jra at samba.org>.
>>>
>>> Can I get a second Team reviewer ?
>> RB+
>>
>> Thanks for catching this!
> What was the consumer in this case?
smbtorture (and from the behaviour it looked like other clients using
the credential stucture would be affected), I was looking in particular
at the winbind.wbclient.wbcLogonUser.wbcLogonUser test and it didn't
work with the user at realm credentials, debugging I was really surprised
to see the 'username' field not filled out as I expected
>
> While very strange, this was deliberate, as it was expected that the
> callers would try and get the principal if that was set at a more
> certain level (eg SPECIFIED compared to GUESS).
not understanding the level stuff mentioned wrt 'specified', 'guess'
etc. but.... if this is the intentional behaviour then so be it, it
seemed like a bug to me (and with my limited understanding this
behaviour seems unnatural) looks like sorry for the noise then if this
is not a bug.

Noel