[PATCH] Fix bug 13465

[Ralph Böhme]

On Mon, Jul 09, 2018 at 04:56:44PM +0200, Ralph Böhme wrote:
>On Sat, Jul 07, 2018 at 02:58:21PM +0200, Ralph Böhme wrote:
>>On Sat, Jul 07, 2018 at 02:48:11PM +0200, Ralph Böhme via samba-technical wrote:
>>>On Sat, Jul 07, 2018 at 01:50:48PM +0200, Ralph Böhme via samba-technical wrote:
>>>>Looking...
>>>
>>>this ones works better. messaging initialisation requires root 
>>>priveleges or out wrappers, neither is given in these python test.
>>>
>>>The new version uses the same approach as smbpasswd. It passed a 
>>>local make test this time, full CI is currently running at:
>>>
>>>https://gitlab.com/samba-team/devel/samba/pipelines/25313916
>>
>>and here's the patch.
>
>sorry, but this one now doesn't work correctly in the clustered case. We really
>need a clustered test env in autobuild.
>
>I'll post a (hopefully) correct patch later on. Amazing how such a simple thing
>can be such a pita. :)

I'm puzzled. If anyone has any idea why this fails in samba.tests.docs with

    fstype
      Expected: NTFS
      Got: 
    ntvfs handler
      Expected: unixuid, default
      Got: 
    printjob username
      Expected: %U
      Got:

Branch: https://gitlab.com/samba-team/devel/samba/commits/slow-autobuild

Full log: https://gitlab.com/samba-team/devel/samba/-/jobs/80523886

I'm basically calling lp_load_initial_only() then messaging_init(), but that has
the above effect of change a few parameter defaults. *scratches head*

I'll continue tomorrow...

-slow

-- 
Ralph Boehme, Samba Team       https://samba.org/
Samba Developer, SerNet GmbH   https://sernet.de/en/samba/
GPG Key Fingerprint:           FAE2 C608 8A24 2520 51C5
                               59E4 AA1E 9B71 2639 9E46