EFW Support

Endian - Security WITHOUT Passion !?

Hello.

Look at this:
w w w .efw-forum.de/www/forum/viewtopic.php?f=16&t=713

Anyone who can translate this to english?

(Google Translate ;) )

Hello people.

I am Endian reseller and currently has more than dissatisfied with the conduct of the company Endian to resellers or customers. Not to mention the quality of the current 2.3 Enterprise or 2.4 community. At mails from customers and resellers, if any, responded only very weak and cryptic. Support Cases will be deported. Even with known bugs that have long been hushed up and be easy. The best example is released 8 months since the 3.2-existing QoS that works to this day is useful or not. There is no clear commitment for or against the community. Even repeatedly offered assistance in testing, development, troubleshooting, be more or less ignored. All together here in the forum to explain exactly would get out of hand and for no purpose.

In short: the current status for me is no longer tenable.

All this has brought me to the decision to found a new fork of the EFW!

Unfortunately, this is not quite so simple and needs the support of a good, functioning community. Therefore my appeal to everyone: MAKE WITH!

The aim should be:

- Highest priority should be a functioning, stable 2.4.
- Currently known bugs need to be addressed shortly.
- All modules mounted on a date (ClamAV, SpamAssassin, Squid, ....)
- SVN repository, documentation of sources
- Wiki, FAQ, bug tracker - with faster response times and properly described bugfixes
- Forum for developers and the community
- Later, an add-on system and new features

Sure some things are missing here. Anyone who has suggestions or wants to participate should contact us by personal message to me.

My humble self is currently in development (programming) might help "nor" not much, since I'm not a programmer. But that can change quickly. I am willing to learn.
However, I can offer a server of the entire project can host. Trouble Shooting and search for bugs, I assume even you. Translation I do already.

So you are all welcome.
Send me your ideas, wishes, suggestions may appear as a Platform / Fork / should.

What tools should be used?
Who can contribute what?
etc.

Would appreciate a successful fork.

P.S. - I'm also very dissapointed about v2.4...
 :(

My appeal to everyone: MAKE WITH!

 :P

i support an endian fork 100%

Hi,

I offer my help as tester.

Thankz

Dear community members,

Endian.com is what it is: a company. Let's face it, they're leveraging on many open source softwares without giving proper means for the community to contribute in an effective manner.

Endian Corp doesn't have an open svn/git repository where developer can peruse and share code or having coding standards and documentation to circumvent the current "obscurity shroud" put in place over the source code, which is taking form of things such as: awkward kernel versions that aren't able (or locked?) to not recognise basic hardware such as embedded usb wireless cards, etc.  Hey Endian Corp... your hotspot functionality doesn't worth all the pain you're impose to everyone in deploying cutted down linux kernel functionalities or weird customized things in such a way that hardware isn't properly detected or running properly (apart of heavily bugged functionalities that still plague 2.4 version...)

Respect toward the community, previous developers & teams (smoothwall coders, etc.) and the open source spirit is beginning with such basic things.  Without respect, how Endian could expect something from community?

IMHO, this is pretty much a closed product now, Exploiting open sources by walking inside a "gray area" dangerously close to goes against GPL licencing's principles/philosophy if it's not against their rules (a potential litigation case for FSF? :-) ) 

So, to all community members being frustrated trying to make something happen with endian, instead of forking, maybe you can check for alternatives such as Untangle, NetCop (they have an hotspot too), pfsense even astaro.

Cheers,

i may be moving to pfsense or astero soon.

And again, we start what is somewhat killing the adoption of Linux, forking.
You linux ppl love to fork, Endian forked from IPCop, PfSense from smoothwall, zillions of forks around ubuntu, even Google!
I think that 10 people working together must produce a better product than 5 groups of 2 people working in parallel about the same issues.
We'll probably repeating the same work!

I completely agree about all the big problems around Endian 2.3. I didn't fully tested 2.4, so I can't say anything.
Lack of support, critical bugs that are not fixed on community, etc. etc..
I needed to invest tons of hours to get my 2.3 ironed. Happily, now it's rock stable for me, and simply works => Boss happy  :D => Big Boss happy  :D => me happy  ;D

Maybe we dont need a fork, just a working group that creates an unofficial 2.4.1 release, with all bugs fixed.
And an open SVN to synchronize with.
Once you clear up some problems, the system is very powerful, more than many commercial ones ($$$$$$$$$)
Agree on all, except on the add-on system. This is the thing I most hate on IPCop.
Everything should be on the main package. If you need it, power on the service. If you don't, leave it off, just like now.
Otherwise there won't be any control, and any new version release could be a hell (it happens to me on IPCop).

I agree with mrkroket

I believe the main problem at the moment is the conspicuous absence of developers who are able or willing to assist with the issues - possibly this is simply a lack of feedback from them.

I would be happy to help out with specific problems - my main issue is lack of assistance with kernel/install issues.

OK.

Let´s do it.

I will setup a server with Redmine h t t p :// w w w .redmine.org/
In my opinion this would be a good platform for developers with all that we need. Also i create a SVN with the available sources from endian.

Think this should be up and running at 13. July 2010.

Then we can discuss at the forum there for all that we need.

Hi guys,

I suggest you try Ipfire, iv'e been testing it and so far its not too bad. But  I still like Endian though, but feature wise its better than endian in some area's.

Still waiting for that fork,  :  )

OK.

It´s up and running....

h t t p : / / e n d i a n.secureit.at

aender

I thinl your will need a register/login page (I cannot see one)

in the right upper corner you find register ..../account/register

Hello all,

I have to completely disagree with mrkroket and the way this "fork" is going..
Please dont compare apples with oranges here. One should look on the reasons people are concidering a fork.
So before we start discussing about IPCop forking to Endian and some Ubuntu forks, lets take a look at the situation here and now with Endian:

- The most important part of an Open Source Project is its community. Where do you see Endian S.r.l. having a community? Where is the communication?

- You very correctly say: How many people of the COMMUNITY do you see working on the project? Let me take a wild guess. 0?

- Do you even know what work Endian S.r.l. is doing behind closed doors? Were were all desperately waiting for a fix to 2.3 when all of a sudden we got a new major release with the same bugs and as a cherry on top, a few more.

- Speaking of Open Source, please read again burja2's post above. Apparently Endian S.r.l. hasnt posted any full source code for two versions now.

- Is there a better example on why a fork is needed? If critical bugs are fixed but Endian S.r.l. does NOT publish them for the community edition, what kind of signal does that give?
You say you spend tons of hours getting 2.3 ironed out. Now imagine you posted all those bugs on Endian's tracked and also provided for the fixes. Do you think they would all end up in the community edition, even if you managed to get hold of someone at Endian S.r.l. to inform them?

- Would it be a REAL Open Source project, with a REAL supported community behind it, I would support that fully. But there is no community, there is no communication between Endian S.r.l. and the people using the community edition. Having a  of unofficial forums here and there, where people using the software try to help each other, does not qualify as an open source community. In an open source community there is a healthy communication between all members working with the same goal, make the software better.

I would fully support a project that helps the PEOPLE. I will not help create a "better" Endian Firewall when the company behind it treats its community like that.

I am happy to see, that more and more Endian users feel exactly the same, as I do.

Yes, Endian is no community project, no matter if Endian names their free version "Community version" or not.
There is no community.

It is just a company named "Endian" who forked IPCop with the goal to steal open source software, close it up and then sell it with a profit. Endian is not open source. There are no open sources, at least not in a way, as the GPL licences say that they have to exist.
What Endian does is illegal. They violate the GPL licences of the software packages that they use for their product.

And there are some people like us, who use the free version and communicate via multiple unofficial forums that we have done in order to help us each other with this software. That is no community, either.

It is time to fork and build a true community project. But is it possible without the sources of 2.4 / only fragments of sources of older versions?

If it is possible, we should organize ourselves and fork, indeed.

But as happy as I am about the action of aender, I do not think that it was a good start.
We should organize ourselves first, find a mission and goals that we all agree to fight for. THEN we should find technical means how to achieve this goals (website, forum, version control, etc.).
It is about humans first, not about technical instruments.
It is about a community. Not about software. That's the point.

So I think that in first place we should make a list of people who are interested in participating, including their skills and how they think that they could help the project.
I can take over this part of organization.

So please:

If you are interested in helping forking Endian, please send an email to:
t-o-m@gmx.net
and give the following information:

1. Your name or nickname
2. An email that we can communicate with you
3. An Skype username, if existing, so that we can chat or make realtime group discussions, if necessary
4. Your relevant skills with level of expertise (1 beginner, 2 intermediate, 3 good, 4 expert, 5 guru)
5. Your amount of time that you will be willing to involve yourself in hours (1 a few hours per month when there is something that I can help, 2 2-3 hours per week, 3 5-8 hours per week, 4 even more)
6. Your idea(s) (if you have) about a nice name for our fork

Examples for relevant skills are:

1. Linux tools & services (e.g. knowledge in one or more of the following programs/technologies: DHCP, DNS, QoS, Proxy, Antivirus, IDS, Software RAID, Hardware RAID, UPS, HTTPD, Networking such as bonding, multi-WAN, hardware monitoring, smartd, VPN, etc., etc., etc... Just look at all the components of Endian and see if you can help in any of them. Or look if there is something that you know and you think it would be great to integrate into the fork!)

2. Web-Development for the GUI

3. Software-Development for Tools such as efw-monitor, or any other software that could enpower our fork, e.g. iPhone apps for monitoring your Endian-forks, etc., etc., etc.

4. Kernel & compiling sources & under the hood: Do you know how to build a kernel in a way that we need it? Do you know under-the-hood things about Linux that you can contribute? Do you know how to compile sources? Have you experiences in forking a project?

5. Security: Are you a network security specialist? Can you advise us about how you think that Endian-fork should be made so that we achieve the highest possible security strenghs?

6. Tester: Do you like to help by installing betas and try them out and analyze them and feedback to the developers?

7. Distribution building: Do you know how to assemble all those parts to become a turnkey-distribution in form of a bootable ISO?

8. Administrator: Do you know how to assemble computers and or to administer them so that we have a testing server?

9. Graphics: Do you know how to make nice graphics and artwork so that our endian-fork gets a nice logo, design, etc in the webinterface, on our webpage, on our CD-labels, etc.?

10. Webmaster: Do you know how to fix us a decent project website?

11. Technical tools: Do you know how to set up e.g. a webforum, a wiki, a version control, a project management software, a communication platform, a torrent for downloading our ISOs, etc. etc. etc.?

12. Software development management: Do you know how to efficiently lead a group of people to develop software in a decent, efficient, painless and nice way?

13. Project lead: Do you have project management skills so that you can lead all those people and groups of people in a way that we truly achieve the goals and the overall mission?

14. Community management & Communications: Do you know how to write nice press release texts, talk to the media, talk to our community and keep them motivated and in love with our project? Do you know how to manage conflicts in a community?

15. Do you like to document features and write a good manual/ wiki?

16. Do you speak some language other then English and like to translate things, such as the GUI, the website, press releases, etc?

17. Do you have money and/or hardware or other resources, that you want to donate? E.g Hardware for a test & development server, special add on cards e.g. RAID controllers (so that we can make the fork support them!), webspace, bandwith, cash, etc.?

18. Do you have other skills that we could need at some point? E.g. legal advisory (especially about GPL, product names, etc., the legal organization form of endian-fork).

This list is just a brainstorm that I did right now. There are surely many other way in which someone could participate. So if you think you could be useful in some way please just send this email!!
After some time when we collected some names, I will contact you via the email to discuss the next steps that we think that we should undertake to get things started.

Let's see what the result will be!

All the best
Tom

And would be in a new project? The forum is filled with requests, but I'm not sure there are many developers around there.

Well, On Sourceforge they supposely put the 2.4 SRPMS (a 500MB file). I used it to create an almost complete Devel machine.
I must warn you that they aren't complete, many source packages are missing, and others don't recompile without modifying them. For example I can't compile VMWare Tools on Endian kernel, because they don't give the sources. But they are sources of PAE kernel, so I can compile on it without problems.
The good thing is that many packages are just tweaked RHEL rpms. Endian sources are in fact modified Red Hat packages plus the Endian code on top of it (efw-*.rpm packages).
So it's viable to create a fork, or at least a Source Repository.

I don't think there are too many... I barely see any modification on this forum except some slight script changes. I hope I'm wrong.

Disclaimer: I am only temporarily using efw because pfsense is not yet supported as a PV DomU in Xen (*BSD kernel crashes), so I am forced to use a *NIX based firewall.

As someone who has relevant skills: why would I put my effort toward forking an essentially closed source system?  It wouldn't be a fork, but a complete rewrite instead.  In addition, why would I want to work on efw when I believe other projects are more featured and less buggy (eg. pfsense and its superb QOS system).

TBH, endian has a really crappy community, the two main reasons being that there is more than one support forum and devs don't participate on the forums.  If you check out other OSS projects, the devs keep in great touch with their base.  Here, everyone is in the dark about how everything works.  I made a post here about how wlan0 isn't detected in the webGUI, and NO ONE responded; on something that basic, either it should work or have a known bug report.  I had to bridge it to another VM and use hostapd, but I see no reason why efw itself couldn't have that incorporated.

I don't mean to say any of this in a negative light, but I would love to hear a convincing argument against the points drawn above.

agreed 100%

if we could at least have some contact with the devs on these forums, that would be a great start.

So I guess we will not see a fork of EFW?

Who even administers these "unofficial" forums?

Feels like only members talking here.

To the above proposal to create a REAL open source fork, I can be contacted for sponsorship.

Interesting! Could you please be a little preciser on what you mean with sponsorship?
Tomakos

Well as far as the attempt of fellow aender, I do not know, but it seems without activity.
As far as for my attempt in gathering some fellow campaigners, unfortunately I have not received a single email up to today. The interest to contribute seems to be minimal.

Tomakos

What kind of sponsorship? Aimed to what?

i support an endian fork 100%

I like the setup form EFW very much. But it has bugs. The only way I can see it, is that the manufacturer is not interested in his own product anymore. The manufacturer can get help from the community, but remains silent.

This situation has some parallels with Openoffice and Libre office. In short: when the manufacturer doesn't improve his product (with or without help from the community), the product is dead. Personally I think that the manufacturer of EFW is not interested in improving EFW anymore because his corporate products won't sell. Or maybe the manufacturer only want's to sell his corporate products (so he want's to dump the community option)? 

For those reasons I would like to see a fork (there are many descriptions for setting up a new product that is based on an inferior product). I'm not a programmer, but I can write and I can test.

I think you were right. I am facing similar query and your post just helps in sorting it out..Thanks for sharing your view.

Finally everyone is waking up that Endian project is just about dead. Screwed customers and the community. 3.0 community is waiting to come out when the cows come flying home.  Move to pfsense and forget these morons. Why create a fork? More effort for a project that was once ok but you are biting off a lot. Just making it 64 bit is only 8 years overdue. It appears they are in a cash crunch and need everyone to update to fill their coffers. I would not mind if it was or good or maintained up to date. I am sorry to say we should let Endian die out and maintain this project.

if endian has had some attentions to community: one single official forum, some devs writing. I don't think that more was necessary. Too bed, it was a very interesting project.

I have a lot of time invested with Endian Community, over 12 firewalls deployed in various state.  Work great but I need a project that is constantly being updated and a large user base.

I have been playing with pfSense the past  of weeks and really like it.  It's a huge learning curve coming from Endian, but much more powerful. 

What other UTMs are you guys moving to? 

I've seen IPFire, it seems interesting.

I think lot of things changed since this discussion started, we learned from our mistakes, for sure the project is far from being dead!
3.0 and now 3.2 improved a lot the quality, the version is now well maintained with constant (monthly) updates.
JIRA bug tracker has been opened to the community to report and discuss about real problems and support/qa/developers are answering and replying officially there.
Weekly we're having a look at this forum to see whether something came up.

Hope this means something.

Cheers,
Luca Giovenzana
QA Manager at Endian

Sounds good. I am trying EFW Community for the first time in years. I contributed a bit to IPCop over the years (not as an official developer) but it's been a while since any releases have been propagated and it's been a good while since any SVN updates have been submitted.

I maintained an adaptation of IPCop to run on old Cobalt servers dubbed Raqcop complete with a bar graph bandwidth meters on the LCD character display plus some changes to the graphs adding hourly as well. I also added a line quality graph that pings the primary DNS server by default but has a file it will look at first if one wants to put their own address they want it to ping to or in case the primary dns is not pingable. I was about halfway done trying to port the Cobalt patch to 3.10 kernels and beyond. 3.4 was the last version that I was able to get it ported over. The kernel has changed so much I'd prefer to use stock or updated third party kernel drivers. The only Cobalt left that would do any justice to todays firewall routers is the 550 with a 1400 PIII dropped into the socket and then there is the noise. Lower rpm fans were a must to tolerate it. My future boxes will be fanless with heatsinks. I'm so over fans wearing out and making noise!

Realizing this is the Community edition, the area I am having the most difficulty with is the urlfilter. Aside from me not getting it to work, I notice it does NOT use squidguard to create databases and the list is not selectable by the user. The best way to use Squidguard is to download the list to /tmp as you do now since it is linked to a TMPFS directory and then let Squidguard create the databases while still in /tmp and then copy that to the proper directory. The database method is much more efficient and less taxing on the proxy than trying to get it to read 18000000 lines of ip addresses directly from text but you want to avoid creating the databases on a hard drive, it takes a lot longer and thrashes the drive so do all that in TMPFS.

There is a lot I really do like too. On the official report/request channel, are patches ok to submit for review?

Hi Dave,

the official place we're mostly working on is the bug tracker jira.endian.com you can open an improvement and attach patches.
We're thinking about opening our git but still work in progress.
For development discussion you can write to efw-devel@

Thank you and cheers,
Luca

Opening the git would be helpful as I like to have a stock source and a modified source where I can test build and run diffs between the two trees and submit for approval.

The bug tracker site though, opened my eyes as to how much development is happening in Endian and I'm encouraged by it. Also submitted a few minor improvements which are in testing, an hourly graph and a decent favicon logo with the EFW Community green logo with a transparent background. I see a lot of activity at the bug tracker site. Very encouraging.

Endian Firewall never had the real Dev sources available.
Source Packages are incomplete, some SRPMS here and there and not a clear way to contribute or recompile things.

Dear all,

Could anybody provide me a sample of vba/ms access/ sql/ office application etc. project.
Actually i m looking this to help graduate students in IT/CS.

Any Help?

rgds,
Aligahk06

Sorry to revive an old thread. I have been looking at the code for Endian 3.3.2 and have been modifying for dynamic zones. I am tired of the limits created by R,G,B,O zones. I hope to completely eradicate the fixed zones and move to dynamic zones and expand DHCP to cover each zone. nothing published yet, but I plan to setup a github spot for the build. I see a lot of strange python redirecting to perl/cgi in this version so I assume Endian plans to move to a python based platform at some point. I will keep the original perl/CGI and get rid of the python since it is barely implemented. I am curious if anyone has done this before i recreate the wheel? I see netdeep has created some type of ipcop/smoothwall for (possibly Endian) but it is still locked into the 4 zones.

Thanks,
Bryan

I've started the fork project - razwall dot com