Slashdot Mirror

First of all, why use crappy openldap when you can use the Netspace directory server that red hat bought and opensourced.

I have foung openLDAP to be reliable, compatible and easy to use. Can you elaborate on why you think it is crap?

There is a reason why they paid 23$ millions for it...

And the reasons are?

Then, AD isn't just a LDAP server with usernames and passwords....

Nor is openLDAP just a store for Windows user names and passwords. I use an openLDAP server for Windows services as well as providing user configuration for other services such as sendmail. The great advantage of using FOSS is that you are free from vendor lock in and can consider non-proprietary alternatives in other areas of your network.

Which is why many people can only use Windows setups. There's nothing like AD in the FOSS world. To start with, FOSS client apps should be lockdown-able from the server. But you can't do that...

I mean, in a office with a linux server and some linux clients, try to lockdown some options on Firefox, the desktop, evolution....surprise, you can't do it. Oh, yeah, there're a lot of workarounds everywhere, but they are different if you use KDE or Gnome or depending on the app you are using. It's a horrible mess.

Nowhere in the article do I see a desire to use FOSS desktop clients. The submitter simply wants to replace AD server with a non MS LDAP based alternative.

Windows clients and servers, on the other hand, are VERY well coupled. The day someone cares to fix this in the FOSS world, a lot of people will start using Linux in corporate networks.

This is otherwise known as vendor lock in. Some of use have tried very hard to break free of it to avoid being held to ransom by a vendor.

Until then, Windows is pretty much the only realistic option. I can't understand why Red Hat, Suse and Ubuntu don't put more efforts on this, it's one of the biggest showstoppers for Linux adoption.

I have been running what you consider an unrealistic option for the best part of a decade. I have yet to be fired. Sirius the consultancy I recommended have a client list of blue chip companines, local govenment and schools. They are all running some form of FOSS backend. You might like to take a fresh look at FOSS, it really works in the real world.

In my previous post I forgot to mention that OGC/Becta are the government agency's responsible for technology in the UK educational environment. It is considerably easier for a UK school to use a Becta accredited supplier than any other supplier. It is an incredible achievement for Sirius to gain that accreditation as no other FOSS consultancy has managed to cut through government red tape thus far.

I think openLDAP should be one of the first products the submitter tries. In my experience it is reliable scalable and free of proprietary cruft. I have used it for years in a commercial network with Samba. OpenLDAP has allowed my company to drastically cut licensing costs, support costs and lengthen hardware lifecycles. As the submitter is UK based I would recommend they contact Sirius. Sirius are the consulting company I use and they are the only UK OGC/Becta accredited FOSS specialist. Sirius have considerable experience in the UK education market and in the submitters position they would be near the top of the list of people to call. Take a look at their client list to see the kind of pedigree they have.

<disclaimer>

I have worked closely with Mark Taylor the CEO of Sirius for a long time now. Please consider anything I say about them biased, contact them youself and make up your own mind about them.

</disclaimer>

RM might be bad, but MS are far worse. They (in the UK will charge schools for installing linux[1]. They, on anti-competitive grounds wont let people know how much MS in schools costs [2]

"This relates to circumstances where schools using Microsoftâ(TM)s School Agreement licensing model, are required to pay Microsoft licensing fees for computers based on Linux, or using OpenOffice.org. Finding ourselves in a position whereby a school pays (say) £169 for a device only to be faced with for example a £30 per year after year payment to Microsoft, for a system that is not running any of their software would just not be acceptable to Becta. Indeed I donâ(TM)t think many people would consider that fair. "

[1] http://www.siriusit.co.uk/myblog/microsoft-tax-on-linux-in-schools-must-end-says-becta.html

[2] http://www.theinquirer.net/gb/inquirer/news/2008/07/08/microsoft-gags-uk-schools

I have just set up a Debian GNU/Linux based DVR system on commodity hardware for a London based construction firm. This was just one machine with eight inputs nowhere near the scale you are working with, however my experiences may be of interest. I used one of the clients existing Pentium 3 fileservers and standard video cameras connected to two IEI IVC-200 capture cards. The base system is Debian testing running motion and apache2.

Some of the replies you have here suggest Zoneminder is suited to a production environment, this is not my experience. Zoneminder is difficult to install, unreliable and over complicated. I'm sure in time the project will mature, however I could not recommend it at this time. Motion offers less functionality but is much easier to install and is reliable. You will have to write you own start and stop scripts as well as web pages to display pictures and video. You will also need to write scripts to periodically archive any video saved on hard disk. None of this should present any problems for a good GNU/Linux administrator. As you seem to have a large number of similar systems you would only need to write one set of scripts and replicate them on each system.

If I were in your situation I would try and use as much of your existing hardware as possible. I assume that you are running a single machine at a variety of remote sites. From a cursory look at the Pelco site the systems you have are standard hardware, and should run GNU/Linux fine. You may find that the capture cards that Pelco provide are not supported so you may need to replace these. If most of your hardware is the same then you can configure just one machine and replicate this on the other machines. Motion supports differing camera resolutions, video/still capture and motion sensing. The motion homepage is at http://www.lavrsen.dk/twiki/bin/view/Motion/WebHom e. If you need strong security GNU/Linux provides you with many options. You can easily verify the suitability of this approach at low cost on a small scale.

If you need any further help contact the company I work for Sirius IT http://www.siriusit.co.uk/. Sirius has good contacts with the free software community and may be able to provide further help.

Steve

We have a number of medium sized enterprises *fully* migrated to Open Source software, and running *way* better on it.
Our best known (in the UK at least) case study is here.
In fact the Group consider Open Source to not merely be a 'substitute' for Microsoft Software, but to have delivered far more real, measurable business benefit than they ever received as a Microsoft Shop.
I am glad Ernie Ball are receiving this great press for their *complete* migration, but they are by no means the first (or the last!) decent-sized enterprise to have done this.

Great idea...

The press release may be found here

Our case study may be found here

We have just finished moving the entire infrastructure of a large UK construction firm over to Linux and Open Source. The TCO gains are enormous and fall into three main categories:

1. Staff. Less maintenance/repairs means staff are freed for productive work (or staff numbers reduced).
2. Licences. Pretty obvious really.
3. Hardware. The upgrade cycle is _drastically_ reduced. Companies depreciate hardware - if you depreciate 300k over three years it costs 100k per year. If your hardware lasts twice as long that goes down to 50k per year.

We have the full case study here and several pages related to TCO at here

We have just finished moving the entire infrastructure of a large UK construction firm over to Linux and Open Source. The TCO gains are enormous and fall into three main categories:

1. Staff. Less maintenance/repairs means staff are freed for productive work (or staff numbers reduced).
2. Licences. Pretty obvious really.
3. Hardware. The upgrade cycle is _drastically_ reduced. Companies depreciate hardware - if you depreciate 300k over three years it costs 100k per year. If your hardware lasts twice as long that goes down to 50k per year.

We have the full case study here and several pages related to TCO at here