ctdb in autobuild broken -- ctdb depends on winbind now????

[Jeremy Allison]

On Sat, Jan 07, 2017 at 05:54:24AM +1100, Martin Schwenke wrote:
> On Fri, 6 Jan 2017 16:00:22 +0100, Volker Lendecke <vl at samba.org> wrote:
> 
> > On Fri, Jan 06, 2017 at 06:25:41PM +1100, Martin Schwenke wrote:
> > > On Fri, 6 Jan 2017 07:59:18 +0100, Volker Lendecke <vl at samba.org> wrote:
> > >   
> > > > On Fri, Jan 06, 2017 at 05:53:56PM +1100, Martin Schwenke wrote:  
> >  [...]  
> >  [...]  
> > > >  [...]    
> >  [...]  
> >  [...]  
> > > > 
> > > > Sure, but as I said: This is a winbind failure, and I don't have the
> > > > chance to take a serious look today. Leaving it broken is also no
> > > > option. So please push the reverts just posted.  
> > > 
> > > As discussed on IRC...
> > > 
> > > Can you please merge in my autobuild branch with the reverts and run a
> > > (or several) private autobuild?  If it passes reliably then:
> > > 
> > > Reviewed-by: Martin Schwenke <martin at meltin.net>  
> > 
> > Survived two autobuilds in a row cleanly. Pushed.
> > 
> > I spent the day trying to find a way to tackle this but failed. If
> > possible, we might find some time at SambaXP to work on the patches
> > together.
> 
> It appears that your autobuild has just failed after both Andreas and
> I got an autobuild through:
> 
>   [1995(13838)/2028 at 3h5m1s] samba4.drs.replica_sync.python(promoted_dc)(promoted_dc:local)
>   UNEXPECTED(error): samba4.drs.replica_sync.python(promoted_dc).replica_sync.DrsReplicaSyncTestCase.test_ReplConflictsRemoteWin_with_child(promoted_dc:local)
>   REASON: Exception: Exception: Traceback (most recent call last):
>     File "/memdisk/vlendec/a/b129423/samba/source4/torture/drs/python/replica_sync.py", line 260, in test_ReplConflictsRemoteWin_with_child
>       self.ou2 = self._create_ou(self.ldb_dc2, "OU=Test Parent Remote Conflict")
>     File "/memdisk/vlendec/a/b129423/samba/source4/torture/drs/python/replica_sync.py", line 95, in _create_ou
>       samdb.add_ldif(ldif)
>     File "bin/python/samba/__init__.py", line 225, in add_ldif
>       self.add(msg, controls)
>   LdbError: (68, 'LDAP error 68 LDAP_ENTRY_ALREADY_EXISTS -  <Entry OU=Test Parent Remote Conflict,DC=samba,DC=example,DC=com already exists> <>')
> 
> I think this is a case of "it's not you, it's me... or just some other
> flakiness".  ;-)

Exactly. I think you and Volker just got unlucky in autobuild
flakyness at the wrong time, precipitating the revert request.

> I think I had 5 failures in a row and I saw several other failures
> along the way before mine and Andreas' finally landed.  Now we're back
> to fail.  Not sure what to make of it...  :-(

Tacking them down, one at a time as always. Metze, you, I
and many others have been doing work on autobuild reliability
for a while now, but it's a sisyphean task. So long as we
get an 80% push-through rate and keep working on the other
20% I think we're OK. It's when the push success rate drops
below 80% or so that people get annoyed.

I think your idea of moving flakey tests to early in the lists
is an excellent one !