Author Topic: Zentyal 4.0 Roadmap Published!  (Read 30294 times)

StuartNaylor

  • Guest
Re: Zentyal 4.0 Roadmap Published!
« Reply #15 on: July 27, 2014, 11:30:37 am »
This is just guess work but Zentyal are carving their niche.

I read about the 4.0 Roadmap and initially was quite despondent.
Yeah all these modules are a loss but if this means the developers and staff can concentrate on a Samba4 / Openchange core then its swings and roundabouts.

I have been a long term protagonist of making Zentyal slimmer and at first I thought wow this has gone too far.

I am starting to look at zentyal not being my gateway any more and actually this isn't detrimental.

I am a sysadmin who tries to provide linux solutions but often by client dictate this means windows desktops.
I don't think this is unusual also I am an ex SBS solution provider.

As far as I am aware Zentyal is the only completely functional AD/Exchange replacement and that is pretty important. So when I think about it, it would seem natural to make that their main goal.

Due to windows being  part of client dictate I need an AD server that is really interoperable and for me Samba4 is extremely important as an authentication server.
Having all manners of authentication support is hugely important and the central part of how I view Zentyal. Then file sharing and even mail is secondary.

That is Zentyal and they could probably also remove the proxy as far as I am concerned. Get the previous right and extensible is so important its the only importance, as with all honesty there has been stability problems.

It doesn't make sense to follow the all in-one philosophy of SBS as hardware is so cheap whilst the cost of maintaining the complexity of an all-in one is so high.

Also I wish Zentyal would open up and make it more extensible. Opensource in isolation from other Opensource is just weaker Opensource. A huge part of the Open of Opensource to me is that it is Open to choice.

The menu system in Zentyal is static and it shouldn't need a developer to add custom menu items. We should be able to add custom menus in a similar fashion to Joomla or Wordpress where a url or script can be accessed.
It is very possible to offer more without support requirements and the ability to allow custom menu's easily, means Zentyal is a more interoperable and better system.

Soon as that is allowed one of the first things I would add is a webmin menu that also contains a start and stop button.
Webmin is as secure as you make it and also any web based admin console carries huge security considerations.
These are considerations though and just like the Zentyal webadmin console how its accessed and who can access it is extremely important.

When it comes to backups and downtime costs I am probably more likely to have a second Zentyal server that will be periodically using rsync to copy user data.
Currently its a bit clunky as the scripts to change home folders and networks shares needs to be run but its a five minute job.

If anybody has ever done an emergency backup and restore after fixing the server fault you will know this is not a nice time.
It takes too long and with modern hardware costs I can easily justify a second backup server with enough power to get me going.

If you want to do the classic backup way then really the backup and restore module in Zentyal really wasn't much cop.
All I can say is install webmin and bacula and have a look at the bacula module in webmin and I think you need to have a look as the zentyal loss might now seem less.

Irrespective of webmin its still a false claim that zentyal doesn't have a backup utility as it has the whole Ubuntu repository full of different backup utilities.
My preference for what I call offsite backup for disaster scenario's is bacula and the webmin module.
Some might want a GUI version and install on the zenbuntu-desktop of the server. I just happen to be in the headless crowd when it comes to servers.
Also even though its not in the Ubuntu repo http://www.diffingo.com/oss/fwbackups fwbackups did get really good reviews for being easy.

I keep banging on about webmin, but it is such a good compliment to Zentyal and its not an alternative but provides some things that zentyal misses.

The bacula module is extremely good, we are talking enterprise grade backup services here.
Monitor again Ubuntu repo mon and webmin has a module, same with UPS.

Saying all this is that I believe there is a misconception in the community that dropping these modules means we lose the ability to have those functions.
What I do believe is that Zentyal irrespective of intentions do not give an impression that in the bigger scheme of things its just an Ubuntu server offering.
Yes it has a great web framework and the developers are leaders in Samba4 and Openchange solutions.
The community can supply and support much of what is needed and this should be more because the community has many heads.

Its self defeating for Zentyal to operate in isolation from other opensource and by embracing them it will create a much bigger product.
Personally I am not really seeing Zentyal as a gateway product any more or at least at this moment in time.

Products like Zentyal and ClearOS are quite heavy as file servers and email servers especially in the manner of Zentyal providing full exchange functionality just scream to me singular server.
I am looking at lighter gateway products that link to my Zentyal server and partition functionality and complexity over the network.
Again name dropping but after a long time with Zentyal I am looking to use http://www.ipfire.org/ with Zentyal. Maybe even http://www.nethserver.org/

Because I see Zentyal being central to authentication I am not keen on them dropping RADIUS as I see that as losing a layer of interoperability.
Again RADIUS is in the Ubuntu repo's so maybe the community can come up with something.

I am presuming the web-server has been dropped so that Zentyal can get rid of the need for a reverse proxy. This is something I am not keen on at all.
I believe the need for a reverse proxy just added to much complexity and it wasn't needed. Is the reverse proxy being dropped and now there is something serving on port 80?
I am stuck and confused though because Zentyal is serving HTTP on the standard ports unless I can base my web applications on Zentyal then I will be forced to use a reverse proxy or non standard ports. For various reasons this is the biggest show stopper for me.

I don't really understand the comments made about IPsec as openVPN can create a secure tunnel so why bother with IPsec?

I am waiting for the first daily builds of 4.0 or maybe further explanation on the whole reverse proxy / web server solution.

The last note is about the user manager and LDAP / AD.

Its about extensibility and I know personally it would be relatively easy to allow custom ldifs in the user manager.
It would be a massive boost to be able to have those fields rather than the samba ldb tools.

Future wise I am hoping that Zentyal will see it can be much more than just an SBS replacement.
https://wiki.samba.org/index.php/Clustered_Samba


   




 
 

   
« Last Edit: July 27, 2014, 12:33:10 pm by StuartNaylor »

CNServices

  • Zen Monk
  • **
  • Posts: 75
  • Karma: +15/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #16 on: July 27, 2014, 05:07:43 pm »
I understand the need to make sure you have a rocksolid foundation for Zentyal.

The complication to this is that there are two kinds of rock solid, from a business point of view. One kind we all share an interest in, the technical reliability kind. The other kind is no less real but harder to define, the feature set reliability kind.

Zentyal has become a drastically moving target. Businesses that prefer the Zentyal they adopted or hoped to adopt are going to think differently about that fact than businesses that are excited about the new Zentyal.

But from a business end user point of view, long term feature set instability is going to be a problem for everybody. The businesses that embrace Zentyal 4.0 can’t reliably predict that they will be able to embrace the next commercial version. There is no credible history for that kind of rock solid.

A business can’t maintain a competitive edge or operational stability if the technology providers keep changing the infrastructure. What small business routinely renovates its physical plant every couple of years? Who wants to do that with their technology infrastructure?

When Debian users start experimenting with a Debian LTS project, I think that says something about a growing negative response to open source rates of change. Business growth generally happens better when the infrastructure is both reliable (technical stability) and predictable (feature set stability).

EDITED to improve clarity.
« Last Edit: July 27, 2014, 05:16:28 pm by CNServices »

ff8jake

  • Zen Apprentice
  • *
  • Posts: 23
  • Karma: +4/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #17 on: July 28, 2014, 05:55:44 pm »
Are the daily builds not available yet, or have I missed the download option somewhere? I plan to eventually deploy this is as an Exchange alternative for approximately 250 users (with varying Outlook versions from 2003 to 2013, and mobile devices), so I would like to help with the bug testing.

Thanks!
« Last Edit: July 28, 2014, 05:59:33 pm by ff8jake »

Torsten73

  • Zen Warrior
  • ***
  • Posts: 174
  • Karma: +6/-1
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #18 on: July 28, 2014, 09:35:46 pm »
@stuart,
mostly i agree with you. But one pint of view more is that we are speaking of a sbs replacement. Furthermore we talked about sbs2003 (thats what i used further) i am not sure in what direction as replacement 4.0 should be.

And we also have an alternative in Univention Corperate Server. Not only in Clear OS. UCS is much more feature rich and extensible. Maybe nearer on a modern Version of Microsofts last SBS as Zentyal will be.
On the other hand, ucs is only free for 5 users. zentyal is totally free.

For me it seems to be, that the development team has been reduced, and the time line is too short for what they want to achive.
So contribute addons is a nice way (like ucs also do) but lets have a look here in the forum. I am not sure, who from us could do this? the activity level in beta testing is really low, i thing for addons it won´t be better.  Lots of new user, and less advanced ones.

I hope one missing thing will comes up, thats the backup of mail store. Could we do this with backula? Backula is also used in UCS.
--------------------------------------------------------------
Zentyal 3.5 (offline) unter Ubuntu12.04.3 YAVDR 0.5 als KVM Host
Action Pack Abo with a running Exc. 2013 :-)

StuartNaylor

  • Guest
Re: Zentyal 4.0 Roadmap Published!
« Reply #19 on: July 29, 2014, 07:19:08 am »
For me it seems to be, that the development team has been reduced, and the time line is too short for what they want to achive.
So contribute addons is a nice way (like ucs also do) but lets have a look here in the forum. I am not sure, who from us could do this? the activity level in beta testing is really low, i thing for addons it won´t be better.  Lots of new user, and less advanced ones.

I hope one missing thing will comes up, thats the backup of mail store. Could we do this with backula? Backula is also used in UCS.

I am really not sure what is happening with the development team, could swear they lost a key person a long time back around 2.0 era.

It has been a strange ride with Zentyal as I am an ex MS developer, mainly I would use MSSQL and bridge applications. Where ever  possible I used off-the-shelf products to create solutions and where ever possible development is kept to a minimal or preferably none. The more advanced you get the bigger ticking time bomb in terms of legacy code you provide. Some IT providers hold clients to ransom this way as they become indispensable. I pride myself in working for my clients best interests.

I have to say it but when it comes to bacula it is a very good example of why Zentyal enforced method of repackaging opensource and only allowing that method is never going to work from my perspective.

http://blog.bacula.org/ almost 3000 downloads each week also it is repackaged and published on my distro of choice Ubuntu.
https://help.ubuntu.com/lts/serverguide/bacula.html so it does get documentation that is offsite also.

Now bacula has a console and from the cli you just use bconsole. Or for the GUI desktop people apt-get install bacula-console-qt
Then its just bat as it needs to be run as root (Bacula Admin Tool)

If your not a GUI desktop fan then there is the webmin module.

Bacula  can backup everything but to be honest I haven't got my head round Openchange. Zarafa used to be a mysql dump.

I am glad you mentioned Bacula as its a perfect example to some questions I have.

Why should I repackage this and create a Zentyal module? Firstly soon as I do I will have to maintain that package.
Then personally I have an absolute hate of the Zentyal redis registry. There isn't a chance that I will hide away configuration settings from /etc to an absolutely pointless windows like registry. Also logically why would create a config method purely to write out a config method.
Then there is the Zentyal object model that consistently seems to change
So I could still do this and create Zentyal buttons that would work from /etc and rebrand Bacula as Zentyal as logically that is the only thing I would be doing.

There is a hugely important side of opensource and that is the many heads of opensource that are user communities. There is a value exchange in opensource that is maximised by number of users and choice. If I package Bacula as Zentyal the user community decreases and also I have negated choice.

It already exists and has large user communities via bacula or the webmin module and this plagiarisation and brand loyalty thing is not why I am here.

Would someone create some community documents on how to use bacula as without doubt it has one of the biggest user bases and that is often a very good indication.

There are many backup utilities and they should all be mentioned as each has its merits.

For me personally the packaging system of Zentyal doesn't sit well with me. It causes work and I shouldn't be creating legacy code that only an expert can do.

A simple menu system with a URL https://192.168.3.1:10000/bacula-backup/ From the screenshot you will see yeah its webmin and that module is contributed by linmin which is a huge part of what I see as the important value exchange between opensource.

Opensource in isolation from other opensource is weaker opensource. What you are asking me to do is in my opinion counter intuitive.

 

 
« Last Edit: July 29, 2014, 08:31:42 am by StuartNaylor »

robb

  • Guest
Re: Zentyal 4.0 Roadmap Published!
« Reply #20 on: July 29, 2014, 10:54:11 am »
Even Zentyal uses a non Zentyal UI with the printer sharing module. It is nothing more than a redirect to the CUPS webinterface.

Anyway, in a sense I have to agree that the continuity suffers a lot with the change from a SBS replacement to a Windows AD/Exchange replacement.

Besides the already mentioned SBS solutions like UCS and ClearOS, you should have a look at Karoshi schoolserver. A webgui put together through a set of bash scripts connecting all kinds of features you ever would want in an all-in-one box (or split over several servers solution)
For now this is fully focussed on a school situation, but imo it shouldn' t be too hard to make a more SBS like variant of it.... (besides that it has a fully automated client install tool for linux clients... )
« Last Edit: July 29, 2014, 10:57:50 am by robb »

StuartNaylor

  • Guest
Re: Zentyal 4.0 Roadmap Published!
« Reply #21 on: July 29, 2014, 11:52:29 am »
Never seen  http://www.karoshi.org.uk/ looks good.

My main argument is that a simple menu manager and extensible ldap would make things much easier to make Zentyal extensible.

Cups is exactly my point as it takes the normal cups package and puts into a zentyal package just for configuration and a button to a url.

Hasn't even been set up with kerberos authentication which you might of thought it would of.

The module making and packaging of zentyal modules is just to much hassle.


yoyomonkey

  • Zen Apprentice
  • *
  • Posts: 1
  • Karma: +1/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #22 on: July 29, 2014, 06:51:33 pm »
I guess I am quite worried about the changes in the 4.0 Roadmap.

I have been using Zentyal for a number of my customers (Solicitors, Warehouse, 1 Independant school) and for me it is a simple one stop shop office infrastructure in a box solution.

With the announcement of removing features like RADIUS and webserver  and Webmail roundcube... I am thinking what will be next?
Granted only 2 of my clients uses roundcube and mail but a large  number of the small start ups i work with work are in rented offices with a rented internet service and use cloud based email services like office 365 or GIACOM and a nearly all of them use the webserver and RADIUS.

They have no need or want to host their internal mail server for a small number of accounts and are happy with using cloud. Only one of the solicitors I am working with now is growing into their own premises and now have budget to bring in email internally from the cloud and have their own elastix solution.

I typically offer solutions like pfsense for firewall RADIUS etc, asterix or elastix for comms, freenas etc initally but a number of these small start ups have very limited budgets and resources they always pick zentyal.

With Zentyal I can design build and test all the features a client needs in a fraction of time keeping costs lower. Every client who is a new starter (Under 10 users)  have always picked zentyal first. I can typically get a majority of what they need built out in Zentyal in a less than a day which in turn kept my costs low. Zentyal allows me to offer my services for free to build an infrastructure for a local charity that needed help.

My clients who are not IT savvy just want to do what they do best and have IT empower them not restrict them. They don't want to understand IT they just want everything to work and zentyal made this easy.

My clients typically want:

Backups
Firewall when a DLINK router just does not cut it :)
Domain Logins
VPN (Life saver for a director on holiday)
Website (Internal/External)
File and Print Share
Messaging
"Cheaper phone Calls"
Zentyal helped me do this.

I am currently about to spend over £400 for the ZECA Zentyal courses and exam but this recent announcement has got rethinking this.

Out of all the small and new business I have come across that are looking for solid IT without paying the earth I have not come across any using outlook 2003 or 2007.

Its bizarre this change as I really thought Zentyal have nailed it with 3.2 for any start up that doesn't have the budgets but want to do things right and cost effectively from the beginning and when the company grows  can easily segment these services from the "one box" when the budgets and requirements expand.

I will give 4.0 daily a test and see how it goes . I could be worrying over nothing here.

Zentyal offers so much and I understand supporting it is a massive demand but I wouldn't mind waiting longer for a new release with "all in it" instead of more frequent releases with less. Look at SME server they take years ! :)



batonac

  • Zen Apprentice
  • *
  • Posts: 14
  • Karma: +0/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #23 on: July 29, 2014, 10:36:34 pm »
Just another small business sysadmin here who has deployed Zentyal, both professional and community, at several business and schools.

I'd like to chime in to say that this roadmap is severely disappointing to me and has me scrambling for alternatives to Zentyal.  I've been able to understand the removal of modules in the past that were used only by a niche market, such as ltsp or kvm, but these changes are very confusing.

What small business doesn't want a built-in backup system on their servers?  Are you expecting everyone to be using a sophisticated visualized environment where data backup is performed outside of the server instance?  If modules like backup, IPS, and UPS, which are huge perks to Zentyal for my usage, are working fine currently, is the development load really too incredibly high to continue hosting them with their current functionality?

I agree that instead of dropping modules like hotcakes Zentyal should seriously consider making everything so modular that "cumbersome" features could be easily developed and maintained by the community.  This would increase both the commercial and community appeal substantially...

idr_mito

  • Zen Apprentice
  • *
  • Posts: 6
  • Karma: +2/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #24 on: July 30, 2014, 10:00:10 pm »
If Zentyal want be a substitute of Windows AD, why don't create two "boxes"?

Zentyal Gateway, and Zentyal Active Exchange

Remove all of these "already created" modules, don't sounds good.
« Last Edit: July 30, 2014, 11:08:47 pm by idr_mito »

mbertens@xs4all.nl

  • Zen Apprentice
  • *
  • Posts: 2
  • Karma: +1/-0
    • View Profile
    • pe2mbs.nl
Re: Zentyal 4.0 Roadmap Published!
« Reply #25 on: July 31, 2014, 08:28:28 am »
So seeing all posts for Zentyal 3.5 and 4.0, I come to the conclusion to stay with 3.4 and never upgrade. And find a replacement for Zentyal.   Its fine with me that Zentyal wants to be a replacement for MS sh*t i dont care, but dropping all kinds of modules Fetchmail, usercorner, web-server, web-mail, UPS monitor and so on... Is to me not a smart choice.

I'm home user (community version) of Zentyal, I have 2 machines; one as gateway/rev.proxy/mail/vpn server and one as data/web server. And i have no windows machines at any more :-D.

When I found Zentyal I was very happy it was/is (until 3.5) what i wanted and needed, and i've recommended Zentyal to others. And i see a lot the same kind of posts, I think that Zentyal should re-think its strategy before only MS personnel like Zentyal, and the rest of the world only sees Zentyal as a secondary MS product.

So will it be "Hasta la vista" Zentyal, or Zentyal "lives another day"

Regards
Marc
Supermirco X7SPA-HF-D525 with RAID1 (system) and RAID5 (data), Firewall, Nginx, WebServer, Mail server, Fetchmail, Webmail, OpenVPN, L2TP/IPSEC VPN, Backup

J. A. Calvo

  • Zentyal Staff
  • Zen Hero
  • *****
  • Posts: 1986
  • Karma: +66/-2
    • View Profile
    • http://blogs.zentyal.org/jacalvo
Re: Zentyal 4.0 Roadmap Published!
« Reply #26 on: July 31, 2014, 12:21:03 pm »
Hi all,

As many of you have raised more or less the same concerns, instead of replying separatily I'm going to try to clarify all your doubts in this post.

First of all, about the dropped modules, as the roadmap announcement says: "Documentation will be provided on configuring some of these services with specific software and naturally, community members interested in maintaining any of these modules will be warmly welcome."

This means that we will write some documentation about suggested specific replacements like Bacula for data backup, etc. Of course community will be very welcome to improve and extend these documentation (as some of you already pointed out).

Also, for people interested in keeping maintaining the modules in the contrib section, we've written already some quick documentation about it:

https://wiki.zentyal.org/wiki/Building_and_maintaining_a_contrib_Zentyal_module

Maybe is not enough but it may be a good starting point and we will be commited to improve it and help anyone with real interest in helping with that.

Please remind that (as some of you were also wondering), Zentyal has been always a modular platform, and even with less officially maintained module, this has not changed at all. Nothing prevents you to build and install existing modules in contrib, or even create your own ones (and of course share them with the rest of the community!).

About suggestions like creating different boxes like "Zentyal Active Exchange" and "Zentyal Gateway", the thing is we don't have infinite resources, and to provide a really good Active Exchange we *need* to focus on that. The work required to maintain modules properly is more than you may think, to give a quick example, is not only about if a module "works" or not (as you have raised the issue that some of the dropped modules just work), you also have to care about them during the upgrades , you need to ensure you don't break them with any change in the core framework, etc. Anyway, if some community members want to fork their own "Zentyal Gateway" distribution with modules like RADIUS, IPS or even new ones, we won't have any problem with that, even we will be willing to help as much as possible.

By the way, about the external mail retrieval (fetchmail), it was removed together with usercorner, but the part that doesn't depend on it is back again in 3.5 now.

Hope this clarifies things a bit. Don't hesitate to ask any other question you may have.
Zentyal Server Lead Developer

Gopher

  • Zen Apprentice
  • *
  • Posts: 45
  • Karma: +7/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #27 on: July 31, 2014, 05:10:14 pm »
OK, I did have a long ranty post ready arguing many of the same points as others already have and how I came to be using Zentyal (I may yet post it). However it seems to me what has happened is Zentyal has made a clear change of direction from an "All-in-One"/"SBS Server" to an AD/Exchange alternative, that is all well and good but what most of us came to Zentyal for was the full package not one new bleeding edge feature that is not as I understand it entirely stable or by any means feature complete. This seems to mean that we should now deploy a separate server alongside Zentyal to deliver the other requisite features or perhaps deploy Zentyal as a virtual instance.
It could be argued that making Zentyal a virtual appliance as standard is actually a very good idea, if it is running critical tasks such as DNS and authentification. Being able to quickly restore a known working up to date image or redeploy it on new hardware after a failure could be a very useful thing. But Zentyal / its community should document a clear, supported procedure and possibly host requirements for creating and maintaining this setup so we don't end up with 1000 threads and questions and 10 different solutions some of which are out of date or never worked in the first place.
I would also suggest the current model of a constantly moving out of support, constantly breaking  moving target "Community" edition and a commercial stable one should be scrapped, there should be ONE current, stable version. A development release should be seperate. Licensing should be for support only, you could even look at support options charged on a pay as you go or event basis. I personally have no particular issue with Zentyal offering chargeable closed-source modules that add "enterprise" features like email archiving that is a legal requirement in some scenarios but not necessarily as easy to implement as it could be.

jonathan38

  • Zen Apprentice
  • *
  • Posts: 4
  • Karma: +0/-0
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #28 on: July 31, 2014, 06:04:22 pm »
So Zentyal is exiting the Firewall/Gateway/UTM market, and focusing purely on Exchange/AD. I wish you guys luck, but I'm pretty disappointed by this.

I've never understood web server, ftp, etc as part of Zentyal. If you have the capabilities to create websites, you don't need a point-click Zentyal interface, your development team can handle that. L7 never really worked right, so that can go. If you're doing OpenChange webmail, dropping WebMail makes sense.

Please correct me if I'm wrong, but Zentyal is the only OpenSource point-click fireware/gateway that offers instant IPS/IDS setup. Zentyal is also the only one that provides point-click IPSec setup. Zentyal is also offers point-click web filtering, again the only open-source based product I can find that does this.

PFsense will do Monitor, UPS, and Backup, but they tend to be squirrelly, and forget IDS/IPS, IPSec support is a running joke, and Content filtering is nearly impossible.

We're two weeks away from our migration from pfsense->zentyal, but I think we're going to reverse that decision after this announcement :(

ctek

  • Zen Warrior
  • ***
  • Posts: 197
  • Karma: +6/-1
    • View Profile
Re: Zentyal 4.0 Roadmap Published!
« Reply #29 on: July 31, 2014, 08:56:09 pm »
I'll share my opinion on the "new" announcement and maybe will help someone see things different.

I've been using Zentyal for some years only at home. I could not make any commercial project with it, I've come across  Zentyal in search of a small SBS 2003 replacement. After years of MS administration i've seen that ebox was doing more than the windows server could do.
As compared to windows you could have one box and do the tasks of three boxes.

However, not having all the features complete to replace a SBS 2003 the customers rejected the proposal of having to do without some features.
Also the fact that the "stable" release was not so stable was another big issue.

After a time i have seen that Zentyal has changed the way they developed the software more to the MS style.
As another user say'd before. It was all in a box and that was great! Now you see yourself without the things that make it great and just a MS clone.
Then the question arises... Why not go with MS since it is the same ? At least the price for a 2012 server is 500$ for 25 clients ? and i can sure create a small mail server on it with programs for windows ?

You can say that: "Ok but it is free".
Correct, but it is not quite free if you think that for each glitch or error encountered the client will pay you to solve it.

You can also say that you still have Open Change that works great with outlook
Correct but does it work "great"? No and there are issues. And Outlook is not the absolute tool. There are other tools used in Enterprises (Ex: Lotus Notes)

Do not get me wrong. I love that Zentyal (Ebox) exists. But in the current form not in what will become.
The strategy that "we want to do it our way and listen to nobody" will sure be the downfall for Zentyal. Adopting the same policy as Ubuntu because of previous success is not a good thing.
You have to listen to the requirements of the consumers / customers, otherwise they will turn on you.

Create software for the customers not for you. Or you will be the only one to use it.
I love Open source and I've use it as much as I can. I agree that the developers should get something in return for the work and time invested in this.
But let's keep it real if something like MS (the principal target of Zentyal ) comes to the market with prices like 500$ or 800$ you can't have any business.

Again: Listen to the consumers / users / customers. They ask and you should provide.
Getting something like Zentyal for free is a great thing and we appreciate this. The people using the free community version in their commercial project should donate a couple of bucks.

Keep in mind that Microsoft has changed and adapted their strategies to beat the linux distros. They are adapting, and  Zentyal is acting in reverse.
drop more of the modules and you will end up just a MS server but with no future.

Think again what were the things that make Zentyal great, where are you going to, and what the people are telling you now...

Best regards
Bogdan