Zentyal Forum, Linux Small Business Server

News and Announcements => News and Announcements => Topic started by: J. A. Calvo on November 21, 2011, 12:04:02 pm

Title: Zentyal 3.0 Roadmap published
Post by: J. A. Calvo on November 21, 2011, 12:04:02 pm
As advanced in the Zentyal Summit few days ago, the main development lines for  Zentyal 3.0 are going to be the upgrade to Ubuntu Server 12.04 LTS, better integration with Ubuntu in general, and the improvements on the Desktop and User experience.

You can find more details at:

http://trac.zentyal.org/milestone/3.0

Note that this is not the definitive roadmap and the community feedback and proposals are really welcome.

The features marked as "extra" are things we want to add but we are not sure yet if we will able to, or in some cases, like the Thin Clients feature, will be done with the support of the community.
Title: Re: Zentyal 3.0 Roadmap published
Post by: Johan_Boshoff on November 21, 2011, 01:51:03 pm
Hi,

I was thinking alot about Zentyal lately and the improvements it needs.  I am really excited about 3.0.

I would really like the option under the FileSystem Backup to backup to more than 1 external disk where a user can just swap hard drives and keep a copy off-site.  The bandwidth in South Africa is really expensive, our customers with 2TB of data (Architects) cannot backup off-site simply because of the costs involved every month.

The users are finding it difficult to swap drives and set the Backup directories etc as it needs some user interaction from them in order to change the drives and for the backup to work properly.

If you would consider this option, I can send through more specific information as to what is needed here.

Thanks, Johan
Title: Re: Zentyal 3.0 Roadmap published
Post by: J. A. Calvo on November 21, 2011, 03:13:51 pm
Thanks Johan, allowing to backup to several destinations (two disks, disk + cloud, etc) is something we have already in our minds. This is just an "small" improvement in the zentyal-ebackup module, and it will probably be included in 3.0, as you can see, the published roadmap does not focus on small improvements, but anyway, your suggestion is really welcome. Thanks!
Title: Re: Zentyal 3.0 Roadmap published
Post by: Sam Graf on November 21, 2011, 03:34:47 pm
It's good to see Zentyal officially committing to an office-in-a-box solution around Ubuntu Server and Desktop. That gives those who lack the freedom (or interest) to migrate to an Ubuntu-centric office solution helpful and welcome clarity about Zentyal's future course.

I personally think it would be helpful to move away from the generic small business server, Windows-server-alternative type of product descriptions as Zentyal consolidates around Ubuntu and the upstream activity associated with it, especially given the vigorous conversation going on around Unity and Ubuntu's apparent interest in the mobile market. Integration around Ubuntu Server and Desktop will open doors where a non-Microsoft boxed office solution is desirable, while presumably making Zentyal increasingly less transparent and "drop-in" ready in other proprietary and open source contexts (given past reluctance to make it increasingly transparent, at any rate). "Ubuntu Office" isn't just an SMB server product, however you look at it :) . Product descriptions would be helpful if they reflected that total office solution concentration.

For what it's worth.
Title: Re: Zentyal 3.0 Roadmap published
Post by: jrugel on December 06, 2011, 08:47:16 pm
What exactly you mean with "User experience: Master-Slave architecture review". Could this review improve (make it easier) the installation in a redundant environment?
Title: Re: Zentyal 3.0 Roadmap published
Post by: exekias on December 07, 2011, 12:37:59 am
What exactly you mean with "User experience: Master-Slave architecture review". Could this review improve (make it easier) the installation in a redundant environment?

Probably,

we want to rethink the current architecture and make it easier to configure. Also we want to get rid of the limitation which forbids module installation on a master :)

Best regards
Title: Re: Zentyal 3.0 Roadmap published
Post by: arun on December 07, 2011, 06:44:43 am
Dear Zentyal Developers,

First of all, congratulation and best of luck, continuously working for the improvement of the product. I found Zentyal the only easy to use server for the SMBs.

1. Combination of Zentyal Server and Zentyal Ubuntu Desktop, will help the administrators, who would like to implement complete integrated opensource setup in his network. In other words it can immediately replace the M$ Client server setup in any SMB.

2. Configuration backup is great, but there should be a remote (within network) incremental backup of data and restore options, because still in may countries bandwidth is still not cheap. I am using rdiffbackup for the purpose.

Again thanks for developing such a nice product.

Arun
Title: Re: Zentyal 3.0 Roadmap published
Post by: Sam Graf on December 07, 2011, 02:56:27 pm
In other words it can immediately replace the M$ Client server setup in any SMB.

I'm going to comment only because Zentyal's course may marginalize significant numbers of SMBs who can't pursue this integrated "office-in-box" solution. I know Zentyal management are comfortable with this, and perhaps the Zentyal community are also comfortable with this. But for some SMBs and their weary technology managers, this essentially means presenting SMBs with a choice among proprietary systems--just because an integrated office solution is built on open source software doesn't mean it isn't for all practical SMB purposes proprietary--rather than encouraging the freedom to leverage FOSS solutions wherever possible.

SMBs who can't build entirely on FOSS offerings are being penalized, even if unintentionally. At my age this comes as no shock, since I have lost the use of dozens of viable alternatives to Microsoft offerings over the decades. But also at my age, this is wearying. I don't understand the either/or mentality at work, when from my perspective it could be both/and.

Of course, as a long time active supporter of open source software (back to my days writing code for Atari computers), I'm very excited about the idea of integrated "drop in" office solution built on FOSS, and I wish Zentyal and its community the very best. I'm deeply disappointed that it effectively forces me and others like me to look elsewhere for a server solution (there is, naturally, the attractive option of forking Zentyal 2.2 and concentrating on a transparent server product).

So all that to say, Zentyal 3.0 won't be a viable solution for "any" SMB, but for those SMBs in a position to embrace a non-Microsoft, Ubuntu-only, semi-proprietary solution. Some will gain more freedom, and some will lose it. Such is life.
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on December 07, 2011, 04:06:09 pm
Sam Graf - this is not true,  zentyal if anything is pushing for a new tightly intergrated alternative to  microsoft windows because of manny vendor locking  stategies from MS.

its not zentyals who is to blame for the problems samba has with microsoft networks and if samba4 will be ready in  zentyal will certainly embrace it...

however we can not be an opensource  ubuntu product while forcing our users to keep using a non opensource desktop...
n'or should we forse them to use tricky hackish solutions to  ducktape-thair desktops to zentyal.

if we want to get any leverage we need our own alternatives to any aspecs of the server / client deal we now have. 
and it will be just that... an alternative 

we should also keep in mind that as long as we will only try to be a windows sbs server clone,  only serving windows clients  we wil always stay behind and become a 2nd rate  software project.   this is not healthy  and it would dry up eventually.

so to quote (but not quote) your own last line:
Zentyal 3.0 won't will be a viable solution for "any" SMB, but even more so than before, for those SMBs in a position to embrace a non-Microsoft, Ubuntu-only, semi-proprietary opensource solution. Some will gain more freedom, and some will not lose it. Such is life.

Title: Re: Zentyal 3.0 Roadmap published
Post by: Sam Graf on December 07, 2011, 04:15:18 pm
Those SMBs who elect To use or must use non-Ubuntu productivity tools face reduced options. It's that simple. It could be different if a both/and approach were adopted, but an either/approach means making SMBs choose between competing systems rather than preserving whatever freedom can be preserved for mixing and matching without treating every non-integrated use cases as a special case requiring custom configuration.

So it is true. I'm not happy about that, but i don't see how it can be anything else. Complete freedom from Microsoft isn't option to every SMB on the planet. It's that simple.
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on December 07, 2011, 04:29:52 pm
i actually dont understand what your are trying to say here,

the roadmap lists  samba4 and kerborose  as the top 2  feature improvemenets for windows networks...
i can not see how you can say that zentyal is going to push away its windows users.

we are not in fact we are only going to offer aditionall support for non windows users.  where we  did not do that before,
if only you take one look at the zentyal desktop package you can see that it has been neglegted for years.  we have been supporting only windows users,  we tried to be a windows clone,  and if anything is true about  what your said earlier the exact oposit would be that...

Zentyal prior to 3.0 wasn't  a viable solution for "any" SMB, but only for those in a position to embrace a Microsoft-only, semi-proprietary solution.

again,  with the zentyal desktop initiative we only want to add (not break away) additional support.   we never mentioned n'or intend, removing any current microsoft windows based features
Title: Re: Zentyal 3.0 Roadmap published
Post by: Sam Graf on December 07, 2011, 04:49:32 pm
I am saying that the desktop initiative has already meant reducing non-Ubuntu solutions (e.g., VDI versus LTSP) to special use status, and that's just in the back room. And that's despite the fact that it would be relatively trivial to support VDI right now without compromising the ability to introduce an LTSP-driven alternative in the future. Why should I think it will be different in vertically integrated mobile and desktop environments?

So if I must have Internet Explorer on the desktop to use a cloud service vital to my business, it is non-trival to commit to a vertically integrated Ubntu-powered total office solution. Or suppose I must use Access to collect complex data to present information to funders, government agencies, or anybody else not embracing an Ubuntu-only office solution? And those are just a simple, real world examples. Things get even more complex as SMBs sort out mobile, remote, and virtual office considerations. Wine doesn't solve all non-Ubuntu business problems any more than LibreOffice does.
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on December 07, 2011, 05:44:05 pm
sam  i can see that you think you have to defend your windows desktop choice,  but i dont see why...  where did i  or anyone else say that where going to reduce functionallity for non ubuntu users.

i really dont have a clue why you seam to have come to the conclusion that zentyal is aproaching this in a  either/or fasion




Title: Re: Zentyal 3.0 Roadmap published
Post by: jsalamero on December 07, 2011, 06:37:56 pm
Hi Sam,

I think you totally misunderstood the Zentyal roadmap. The goal is to implement on server side Single Sign On and Domain Controller for both operating systems Windows (with Samba4) and Linux (with Kerberos). Actually, with Samba4 you get a full drop-in Microsoft Domain Controller replacement (advanced policies, etc.), something that is not present on the Linux world. That's one the things will implement Zentyal Desktop for Linux. To give you an example, on Zentyal server you will able to define a policy: "apply this wallpaper to all the workstations", on Windows will be implemented via policy with Samba4, on Linux will be implemented via Zentyal Desktop.

Zentyal Desktop will preconfigure some applications, do some monitoring, alerts, etc. Will be available for both Windows and Linux, with different features (policies are already implemented on Windows, not on Linux), will ease the configuration and management of desktops. If the desktop is VDI, Thin Client (LTSP, TCOS, etc), Fat Client or bare metal, doesn't mind.

About VDI which seems you are very interested into, I suggested you some time ago how to put some configuration snippets inside /etc/zentyal/dhcp so you could integrate these options you need for VDI with the Zentyal interface management. Give it a try. If VDI become a common deployment in the community, I'm completely sure that these options will come to the web interface too!
Title: Re: Zentyal 3.0 Roadmap published
Post by: Sam Graf on December 07, 2011, 07:21:32 pm
Hi Sam,

I think you totally misunderstood the Zentyal roadmap.

Very possibly. And I'll be happy to admit that. :)

If that's true, then the confusion is rooted in the nature of the Zentyal desktop proposal, as near as I can tell. For example, it seems to me that you are describing something different than I've heard before--something more like what Zentyal Desktop is like now rather than as a Zentyal-branded Ubuntu desktop serving as part of a total office solution, as described (I thought), by everybody else.

As for VDI, that example really means nothing as far as Zentyal goes outside the LTSP-is-to-be-preferred conversation. If Zentyal is willing to entertain LTSP support as a core feature, why not the easy-to-implement, relatively speaking, support for already existing DHCP functionality? Can you see how all this might read as putting non-Ubuntu solutions, broadly speaking, into special use case as a matter of general policy? Especially if the condition for core inclusion is widespread community support in a Linux-oriented community?

Consider the UPS support conversation instead, to understand how easy it is for me to get lost by Zentyal roadmap priorities. I don't mean that as a criticism, but simply as an example of conversation about priorities I cannot grasp. No doubt the limitation is with me.

Perhaps the real problem is just one of timing in my case alone. While I've tried to keep my thoughts in line with a broad category of small business and non-profit operations, maybe I've missed something important along the way. In my world of rural Michigan, there are people who are so technically challenged that they see technology as little more than fancy typewrites. The rest are still technically challenged, but they are young and have a vision, driven largely by their consumer experience, of technology without walls--so-called digital natives. Us digital immigrants seem to them to lack vision. I've been trying to lobby for a Zentyal that endorses their wall-less vision, since it's those people that will replace me very much sooner than later. It's very possible that Zentyal has done just that and I completely missed it. My very sincere apologies if that's the case.
Title: Re: Zentyal 3.0 Roadmap published
Post by: arun on December 08, 2011, 01:38:17 pm
As I am already trying to use Zentyal Desktop in production environment with Zentyal 2.0.
Will there be development version released for Zentyal Desktop  ???

Arun
Title: Re: Zentyal 3.0 Roadmap published
Post by: jsalamero on December 08, 2011, 03:05:15 pm
Quote
If that's true, then the confusion is rooted in the nature of the Zentyal desktop proposal, as near as I can tell. For example, it seems to me that you are describing something different than I've heard before--something more like what Zentyal Desktop is like now rather than as a Zentyal-branded Ubuntu desktop serving as part of a total office solution, as described (I thought), by everybody else.

Zentyal Desktop will be a set of packages to configure and connect to the server Windows and Ubuntu desktop, no matter if they are bare metal, VDI or LTSP. Will configure authentication, policies, default applications between other things.

Quote
As for VDI, that example really means nothing as far as Zentyal goes outside the LTSP-is-to-be-preferred conversation. If Zentyal is willing to entertain LTSP support as a core feature, why not the easy-to-implement, relatively speaking, support for already existing DHCP functionality? Can you see how all this might read as putting non-Ubuntu solutions, broadly speaking, into special use case as a matter of general policy? Especially if the condition for core inclusion is widespread community support in a Linux-oriented community?

In the forums seems to be a lot of people interested on LTSP, they are working on create some step by step very easy to follow documentation so the technology gets spread withing Zentyal administrators. Eventually this could become a module. If you are very interested on VDI, the way to push this upstream would be the same: create documentation, use cases, describe advantages over other solutions, you can create all this on the wiki. Windows and Ubuntu clients will be both first class supported clients, we are not going to leave behind Windows at any moment. Currently Ubuntu lacks some features they are implemented in a native way on Windows, we want to push that on the desktop side. The same way that we want to push Samba4 to the server becomes a full featured Domain Controller like Windows Server is.

Quote
Consider the UPS support conversation instead, to understand how easy it is for me to get lost by Zentyal roadmap priorities. I don't mean that as a criticism, but simply as an example of conversation about priorities I cannot grasp. No doubt the limitation is with me.

We know that UPS support is a highly demanded module, let's see what we can do here.
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on December 08, 2011, 06:59:30 pm
@sam - now i do understand your point but there are various degrees and opions on how the dekstop integration should evolve.. some people seem to want it an ltp option only others seem to want to go in other directions

it is and will be the primary task of zentyal is to add (with minimal effort if possible)  features that will streamline the incorporation  of a well featured, easily manageble and seamlessly intergrated desktop based on free software.

 use of these additional packages and scripts to build a zentyal-flavored 'semi-official'  install image.
would not take more than a few unhappy hours fideling with remastersys (or any other tool like it) and im sure that it will help new users in many ways (as will it help us (the community to support them).   yet still this is no more than a very small sidestep,  and could be done by just about anyone who ever  build his own remastered disk before..

and i believe i have mentioned enought times that these packages need to be transparrent enought to be usefull in any install case  as a desktop as well as an ltsp-root. or any other possible use-case.
Title: Re: Zentyal 3.0 Roadmap published
Post by: jclambert on January 15, 2012, 12:21:51 am
AD domain replacement is my goal.
 
I have tested some samba based solutions.
... ClearOS
... Zentyal
... Resara

 Resara is samba4 already, and as such a bit ahead as a domain server.  Now, Resara does nothing aside from AD emulation, unlike the other products.  That said, I may be forced down that road if the other two options are not viable by mid year.
Title: Re: Zentyal 3.0 Roadmap published
Post by: robb on January 15, 2012, 09:41:16 am
Zentyal 3.0 will probably be released somewhere in the neighbourhood when ubuntu 12.10 comes out (about October 2012 give or take a month) I'm convinced it will be the best release ever! There will be a lot of exciting new features, with Samba4 and Zentyal Desktop as the major features.
Title: Re: Zentyal 3.0 Roadmap published
Post by: J. A. Calvo on January 15, 2012, 11:54:42 am
The final release will be in September 2012, and yes, let's hope it will be the best release ever! ;)
Title: Re: Zentyal 3.0 Roadmap published
Post by: BuckC on January 24, 2012, 11:00:59 pm
Speaking from someone in the US, many SOHOs, SMBs do not have the $ to upgrade/license Windows servers and/or clients. Many people I speak to are actually looking for a M$ alternative and the license/CAL costs. I am also one who wants a domain controller replacement and have tried other platforms mentioned in this post. Since I am a M$ admin during the day, I have experience in what M$ offers the customer (Enterprise license), so finding a replacement in the Open Source world was hard to find. I currently admin a couple of Zentyal 2.2 installations, and Zentyal is by far the easiest to setup and configure, although I do not use all the features available.

But that is the great thing about this Zentyal solution, you only install/use what you need.

To me Zentyal 3.0 Desktop piece is the finishing touch to a great product, making a somewhat manual complex process into an automated one.

No, I do not work for, nor am I affiliated with Zentyal the company in any manner.
Title: Re: Zentyal 3.0 Roadmap published
Post by: Ken on February 07, 2012, 04:38:04 pm
Please be aware the Ubuntu 12.04 LTS has a bug which affects LDAP.
https://bugs.launchpad.net/bugs/926350
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on February 07, 2012, 09:55:07 pm
dear ken - thank you for mentioning it, but as you may know ubuntu 12.04 is still in a beta state and sutch bugs are to be expected... these should be fixed before conical ships ubuntu ...

Title: Re: Zentyal 3.0 Roadmap published
Post by: Ken on February 11, 2012, 08:45:28 pm
dear ichat, I'm afraid that the issue is not one of technology so much but of policy.  I posted a known fix into a PPA but it causes software license issues.  Unless the license issues are resolved there is a good chance 12.04 will ship with the bug in this report
https://bugs.launchpad.net/bugs/926350

The issue is discussed here:
http://lists.debian.org/debian-legal/2011/02/msg00012.html
    Subject: Re: LGPL library using only LGPL-parts of partially GPL shared library (gnutls, nettle)
    From: Andreas Metzler <ametzler@downhill.at.eu.org>
    Date: Sun, 20 Feb 2011 11:23:37 +0100

    ... Nettle's public key library (libhogweed) uses and links against
    libgmp, which is LGPLv3+. Therefore switching gnutls from gcrypt to
    nettle would break GPLv2-compatibility (GPLv2 without the "or any
    later version " clause). ...
Title: Re: Zentyal 3.0 Roadmap published
Post by: Ken on February 11, 2012, 09:05:36 pm
Some explanations:

With Ubuntu 10.04 (through 11.10)

pam_ldap.so is linked to libgnutls26 which is linked to libgcrypt11

If you log in using an ldap user:
* sudo will work if the ldap connection is non-encrypted (does not use the libgcrypt code which drops setuid)
* sudo will work if the ldap connection is encrypted plus you are running nscd (I'm not really sure why this works, I haven't taken the time to study it)
* sudo will not work if the ldap connection is encrypted and you are not running nscd

With Ubuntu 12.04

pam_ldap.so is linked to libgnutls26 which is still linked to libgcrypt11 (but it's an updated libgnutls26 which could be recompiled to not use libgcrypt11)

* sudo will not work if the ldap connection is encrypted regardless of nscd running or not.
* sudo will work if the ldap connection is not encrypted.

sudo is not the only program, /usr/bin/at has the same issue, basically all setuid (http://en.wikipedia.org/wiki/Setuid) binaries do.

When I rebuild gnutls26 without the configuration option to use libgcrypt11 it switches to using libnettle4 but that is linked against libhogweed which is linked against libgmp which is licensed under the Lesser GPL version 3. Debian does not want their main repository to contain any LGPLv3 code.

There was an alternative fix which is to delete a call to the initialization code in libgcrypt11 at the point where thread callbacks entered the libgcrypt11 library but the full effects of making that change were not understood well enough to consider it harmless.  It was proposed that a new API with a second thread callback entry point that did not drop setuid would be added but that was rejected on the grounds that they were going to switch GnuTLS to stop using libgcrypt11 and use nettle instead.  See
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=566351#112

Other Linux distributions avoid all of this by linking pam_ldap.so against libnss3 instead of gnutls.
Title: Re: Zentyal 3.0 Roadmap published
Post by: nvarona on March 02, 2012, 01:32:42 am
Hello,

Webserver Module for Zentyal is good and serving will but sometimes here comes the problem

Is Zentyal Developers Team has a plan to add at NGINX webserver as a next choice for webserver module. Not a replacement but a second choices for webserver.

Thanks.
Title: Re: Zentyal 3.0 Roadmap published
Post by: graylion on May 17, 2012, 01:26:37 am
https://bugs.launchpad.net/ubuntu/+source/sudo/+bug/423252

bug got fixed today it seems
Title: Re: Zentyal 3.0 Roadmap published
Post by: wlaszi on August 05, 2012, 10:11:49 am
I know, maybe I'm late, but here is my wishlist:
- an OCSP service or a built-in CRL upload method to refresh my external (vpn) servers CRLs when I revoke a cert.
- advanced cert settings interface like XCA
- CA-LDAP integration: link cert to user/machine from LDAP with one click and pre-fill cert create form with the available ldap values.
- less apache dependency, support nginx
- built-in common contact/addressbook support
- pdf manual. Use a public script  to create it myself from html is an unnecesarry round for everybody.
Title: Re: Zentyal 3.0 Roadmap published
Post by: ichat on August 05, 2012, 03:24:42 pm
your not maybe late you are really late... 

zentyal is already in first feature freeze release, that many no new functions will be added   therefor the next version will most likely be the first  feature complete version...   and hopefully soon after some additonal bugs will be fixed and  we might be off to start  the final stage of bugsquashing...

as things look now,  the community based discussion about zentyals  3.2 roadmap will hopefully start in the next  couple of weeks with its firts innitial discovery steps.

as some of your feature request have been submitted in the right format some time back 1 or more may even be incoperated in the next release version.  (But absolutely no promisses here).
for example some  ca-cert released feature requests, and replacing apache with some other httpd (probably not nginex though because  lighthttpd  outperforms nginx  by a mile or so on  dynamically rendered pages ...  (nginx is mutch more a high load static pages  server (reverse proxy)

yet some have already been implemented ..  like your central contact database  is allready part of the zarafa groupware suite that is included in zentyals main track so there isn't any need to add it anywhere else is there? 

if you would like to post 'detailed' feature request please do so  in the right forum  etc..
Title: Re: Zentyal 3.0 Roadmap published
Post by: wlaszi on August 05, 2012, 07:32:07 pm
yet some have already been implemented ..  like your central contact database  is allready part of the zarafa groupware suite that is included in zentyals main track so there isn't any need to add it anywhere else is there? 
I know, I'm late but if I don't post it now somewhere, I'll forget it as usual :)

The built-in central contact database for Zarafa(!) is available but need to be extend it to use with email client (I guess this is the main use). Now, it's possible only with lot of tricks (http://trac.zentyal.org/wiki/Documentation/Community/HowTo/ZarafaSetupWithOutlookThunderbirdSync). :(