Joining to domain incomplete: Cannot execute "nsupdate" server says timed out

[aalvaro23]

Hello:
Zentyal Masters:
In the final step of joining a zentyal server as aditional DC of windows server 2008 an error comes up related with DNS module.
I think that for a reason unknow (for me) the DNS module of Zentyal is refusing the nsupdate command with the instructions inside the "hY8bKa7Etz" file in  /var/lib/zentyal/tmp/
Log:
EBox::Samba::__ANON__- Joining to domain 'dominio.lan' as DC
EBox::Samba::__ANON__- Trying to contact 'winserver.dominio.lan'
EBox::Samba::__ANON__- Trying to get a kerberos tickect for principal 'administrador@DOMINIO.LAN'
EBox::Samba::__ANON__- Executing domain join
EBox::Module::Base::save - Restarting service for module: dns
EBox::DNS::_launchNSupdate - Cannot contact with named, trying in posthook
EBox::Sudo::_rootError - root command nsupdate -l -t 10 /var/lib/zentyal/tmp/hY8bKa7Etz failed
Error output: ; Communication with server failed: timed out


And off course the module File Sharing won't start. But the zentyal did register in windows as an additional dc because i see him inside Domain Controllers OU in Active Directory, please if someone had resolved this problem or it understand better, a little help will be extrawelcome
 

[aalvaro23]

And , speaking of the same subject, I have something to say about Zentyal 3.0x, I don´t understand the logic of the zentyal developers team at the time of design this new architecture with kerberos and samba4 because rather than make closer the integration with Windows AD(asuming that this is the intention) they implemented a product that cause a lot of problems to the admins that want to integrate zentyal with microsoft platforms(me included). Example: Windows Server 2003, there is NO WAY to integrate zentyal as an additional DC following this so called ¨intuitive logic¨ that always had been present in zentyal, and not either following the official documentation  (or tracs, with the examples), this joining process it only appears to work (hardly) with Windows Server 2008 and 2012, and this is a  REAL COUNTERPRODUCTIVE PROBLEM. If  I´m wrong  please tell me.

[aalvaro23]

Well, seems that so far, I have reason in my posts, but please, the zentyals monsters have to react or anybody seems not to care about this subject. Come on...

Best regards

[Sam Graf]

I'm not sure what reaction you are looking for. There are known issues surrounding Zentyal's integration of Samba 4. The developers have been working on them since 3.0 release, including working with the Samba team. In the meantime, there are some who consider Zentyal not entirely stable in services dependent on Samba 4.

So these things have been publicly discussed and acknowleged already. Bug reports have been filed and closed and in some cases reopened and closed more than once. It's a long way from a case of people not caring. The reality is that Zentyal 3.0 has almost become a work in progress in certain features because the developers are trying to solve the problems. It's also possible that not all problems will be solved with complete satisfaction within the 3.0 architecture but will find better resolution in changes being considered for 3.2. Time will tell.

[aalvaro23]

Sam Graf

The reaction of which I spoke was that precisely, a statement regarding this architecture related with samba4, I think that the desicion was not incorrect, on the contrary is expected that zentyal every day gets closer to the integration with existent platforms that is one of the most powerfull advantage, but the mistake here is to declare as stable this versions that inlcude a defective file sharing module(only speaking in terms of windows integration) so it would have been better to perfect synchronization with AD and go to work in samba 4 easy to achieve a integration not so complicated. The others modules, OK

[Sam Graf]

Sure. I don't disagree with your view on stability. Again, the developers (along with Zentyal marketing) publicly stated that they intentionally moved to Samba 4 even while it was still in beta with the goal of replacing SBS and that they have no regrets about taking that step. Integration with existing infrastructure wasn't a big target, and I'm assuming that they had more confidence in the stability of 3.0 than it proved to have.

There are several potentially controversial things about Zentyal's direction. Given the rapid change happening in small business IT, I doubt it can be otherwise. Zentyal is going to pick its market. In any case, those decisions aren't made primarily at the community level; they are primarily business decisions rather than technical decisions.

All of which is just to say that my interpretation of things is that Samba 4 integration for 3.0 was driven by factors not obvious to the community. I don't have inside information, just what I see in public like everybody else, but that's my interpretation. We have significant endorsements (from places where perhaps Microsoft is not a primary influence) of Zentyal in conjunction with the decision to take a more aggressive approach to Samba 4 integration than might have been expected from a technical point of view. It is probable that Zentyal is moving closer to the Ubuntu business model in terms of aggressive entry into relatively narrow markets.

It'll all be interesting to watch.