samldb and associated patches

[Matthias Dieter Wallnöfer]

Hi Kamen,

Kamen Mazdrashki wrote:
> Hey Matthias,
>
> On Thu, Sep 30, 2010 at 09:53, Matthias Dieter Wallnöfer 
> <mdw at samba.org <mailto:mdw at samba.org>> wrote:
>
>     Hi Kamen & Andrew,
>
>     Kamen Mazdrashki wrote:
>
>         I think it looks good now.
>         I am only little bit concerned about changes in made in
>         source4/dsdb/kcc/kcc_deleted.c
>         <http://gitweb.samba.org/?p=mdw/samba.git;a=blob;f=source4/dsdb/kcc/kcc_deleted.c;h=1726fa4f318a2f18893c6866ef05274bec7533af;hb=8525a44fabcb118ccd5a527e9b9eadb045894df7>:
>
>
>         http://gitweb.samba.org/?p=mdw/samba.git;a=commitdiff;h=8525a44fabcb118ccd5a527e9b9eadb045894df7#patch2
>         But kcc_deleted is just a basic implementation anyway.
>
>     well I really wanted to make it to see all deleted objects.
>
> yep, you are right. What I meant is that kcc_deleted() is going to 
> apply now the same logic
> for all states of the object, and there may be a difference between 
> Tombstone-lifetime
> and DeletedObject-Lifetime values. Anyway, your fix will allow us to 
> have the chance
> for clean all kind of isDeleted objects, thanks :)
Yeah - I think you could look into this later - I really don't 
understand much in this issue.
>
>
>         Regarding "nTDSConnection" object -> kcc service creates them
>         also. You may want to
>         peek kcc implementation how it is done.
>         I think that "Active Directory Sites and Services Manager" is
>         also able to create
>         Connection objects, but I am not sure how it does this.
>
>     No, I speak from the "nTDSDSA" objects (NTDS Settings). The other
>     ones you can create with any trouble, but you need the server and
>     nTDSDSA object already there.
>
> I think this may be achieved with DsAddEntry() drs call?
That is possible
>
>
>     But I think my tests should now be enough to test all interesting
>     scenarios.
>
> I still don't see a test for a subtree of objects with mixed 
> systemFlags :)
> Anyway, I may implement such a test when I start implementing a test
> for DsAddEntry() (I think we still don't have a sane test for this 
> entry point)
Well we could add afterwards one.
>
>
>     So, Kamen & Andrew, if you like my work I hope that it will be
>     merged soon.
>
> IMHO, it is ok to be pushed. I would suggest though for you to wait 
> until the
> end of the week - so that guys at plugfest to get the chance to test only
> what related patches. What do you think?
Okay, I can wait until what you say me: Saturday, Sunday?
> Thanks a lot for your patience Matthias - I am not a fast reviewer, 
> sorry for that!
>
> Oh, I have a patch for your problem with timeouts in FSMO test:
> http://git.samba.org/?p=kamenim/samba.git;a=commitdiff;h=10b472fc739854dcc85c9d3dfc81ca4455ef0759
> Current timeout is about 20 secs and it should be quite enough, even 
> for busy/slow machines.
Very appreciated, thanks.

Matthias