opensuse and suse

I was just wondering if opensuse and suse relationship is the same as Fedora and RHEL, I was just curious because I know that RHEL has put some software into Fedora that from time to time has made Fedora have a very buggy feel to it which I understand that is RHEL test project. Does Suse do the same to openSuse?

Here’s one way of thinking of the chain of processes:

There is a raw cutting edge environment called “Factory” where new, not necessary stable, packages are tested and debugged so they can work in the openSUSE environment. Then there is a front line stable distro named openSUSE where the tested and stable packages from “Factory” are integrated into the next stable public release (called openSUSE). From time to time aspects of openSUSE have a buggy feel because the integration of the new stable packages is not always smooth. Once packages are fully integrated and are running smoothly in a release of openSUSE, those packages can then find their way into the enterprise level products (SLES and SLED). So it’s really a multi-tier process from Factory through the public product openSUSE to the final level, the Enterprise level.

And you can get the story straight from the Lizards mouth here: Linux OS | SUSE Linux Enterprise

ThankYou,

On 01/01/2012 05:06 AM, swerdna wrote:
>
> Here’s one way of thinking of the chain of processes:

a GREAT answer, which i intend to point to for all subsequent similar
questions…


DD http://tinyurl.com/DD-Caveat
openSUSE®, the “German Engineered Automobiles” of operating systems!

Swerdna wrote:

Once packages are fully integrated and are running smoothly in a release of openSUSE, those packages can then find their way into the enterprise level products (SLES and SLED)

So then, seeing the handling of KDE4 and Gnome3, does this mean they are still on KDE3.5 and Gnome2?? :O(you can go for irony or sarcasm here).
And if Factory is what it is, what is Tumbleweed? How does it fit? And if people use tumbleweed, what does it mean to test an alpha or beta? And how can it be meaningful to enhance stability though user interaction, putting out a kmail2 package with an alpha status (maybe not of code, but surely of integration). This is equivalent of saying that the openSUSE distribution is (in its actual version e.g. 12.1) not for production machines. I cannot really see this kind of statement as desired.
If I could agree on very long lines with what Swerdna wrote, I see a huge disorder in the overall project logical setup and this is IMO the reason for this kind of questions.

@Denver: you may point the people asking here, but there is a reason for them asking being puzzled by what they see, and, i am afraid, this reason will not go away, all to the contrary it will swiftly grow.
As usual YMMV.

For my ‘take’ on this, … Tumbleweed was added later.

A number of people did not want to have to wait 8 to 9 months for some of the newer more cutting edge pacakges.

A wish of these people was that like some other distributions, that openSUSE would also have a way of implementing a rolling release, with some of the newer stable packages included. With this and other considerations in mind, one of the SuSE-GmbH packagers (I believe) volunteered to take on the task to lead a rolling release for openSUSE called ‘Tumbleweed’ which would compliment the existing ‘factory’ and ‘nominal’ 8 or 9 month release cycle for a standard openSUSE release. ie once an application was submitted to factory, and working reasonably well in factory, it could then be also moved to Tumbleweed. That is by no means a ‘given’ that a factory package will be moved, but it is an essential criteria for one to be moved. And note also that ‘Tumbleweed’ by no means gets the same level of testing that the nominal milesetone/beta release cycles get.

This had the advantage over ‘just’ going to factory one’s self, that there was an extra quality gate of this SuSE-GmbH packagers view as to the stability of the factory package (before the package went into Tumbleweed)

Means to whom ? Testing the milestone and beta releases are absolutely essential as THEY (and not Tumbleweed) are what form the eventual next release. Tumbleweed, does thou, help in a ‘parallel’ path provide a look as to what some of the problems may be with a factory implementation.

I don’t see that at all.

OP asked a simple question and I gave a simple answer. I intended to leave out the history of Linux as unnecessarily confusing. You went somewhere in between with a cloud of irrelevant side issues that explained nothing.

I see Tumbleweed in conflict with alpha and beta testing. The reason is that people using Tumbleweed are IMO by the very motivation risk loving and potential beta or alpha testers … that now will not test. That said, I am not knowledgeable of how much of the Tumbleweed experience does bleed back into the alpha and beta process. Still I see a conflict.

Evergreen: is conflicting with commercial interest of SUSE. It does however represent an asset for the distribution and does contribute to the picture of reliability of the distribution. It is IMO dangerous to say: openSUSE comes from open = buggy and SUSE as professionally and commercially maintained = professional. Although this is a subtle message and may well be opposed by saying: no tendency at all, IMO I do see this tendency currently.

Financing: as long as the project will have one only sponsor, it will be subject to this. I am aware of the discussion on this, as it is a difficult topic I would not wish to raise this here further. Still it seems to me that these are all very important points that denote a logical mismatch (call it disorder, the magnitude I leave it to personal evaluation. Huge may be too much, agreed upon).
YMMV of course.

I’ll go only by my own experience.

I suppose I am risk averse. I thought about trying Tumbleweed, and decided that I didn’t want the hassle.

Nevertheless, I did try the milestones, the beta and the release candidate for 12.1.

Does that make me a counter-example to your thesis?

I tested milestones by installing on a separate partition of my laptop. Then, where possible, I would boot that partition each day for some testing. Meantime, I could get any real work done with the stable release of 11.4

Later on, as experience showed no major killer defects in 12.1, I installed that on the main linux partition of the laptop. I did not install on the main partition of my primary desktop until the final release.

I will probably do something similar for the 12.2 development cycle.

If I had gone with Tumbleweed, I probably would have installed only KDE. But, testing the milestones on a second partition, I also installed gnome, and I spent some of my testing time with that.

If I had gone with Tumbleweed, I probably would have installed only KDE. But, testing the milestones on a second partition, I also installed gnome, and I spent some of my testing time with that.
In a certain sense, this confirms my doubts.
The thing about an independent partition interest me, but I would not boot often if I cannot share the home. And if I share the home I may run into a configuration problem. How do you address this? BTW, although I am running a 500 GB disk, incredibly space is never enough. So space is also an problem I see with this, but the idea is intriguing me quite some time now. Although, to be meaningful, testing should be done with day by day working practice when the real bugs come out. No?

I don’t think, just because one has Tumbleweed, means one will not test a milestone or beta release.

My main PCs typically lag the current openSUSE by a version. BUT I do have two PCs (one sandbox, and one our GNU/Linux Club’s PC) running Tumbleweed. AND I also test every milestone/beta release. What does that make me ? By the logic in your post, I should have stopped testing milestones/betas. I did not. In fact, I probably test more now.

And in addition, I know other’s who use Tumbleweed who also still install milestones/beta releases. … What does that tell me ? It tells me your point, while logical on the surface, is not what is happening in practise. In practise many users who use Tumbleweed, also still test the milestone and Beta releases.

Also, do not ignore the parallel synergy that comes from factory packages, packaged into Tumbleweed, being tested. I would venture an opinion that advantage of more Tumbleweed testing of factory more than outweighs any loss (which I dispute) from users who have left milestones/beta for Tumbleweed.

Evergreen is community driven. The possibility of trying to do something longer lasting with SLED/SLES was from what I recall discussed, but SLED/SLES were going in a different direction than that of Evergreen. This was discussed already some time back in the early days of Evergreen creation within the community, and also with Novell. Ergo Evergreen fills a niche in the ‘SuSE’ GNU/Linux distribution not met by SLED/SLES, nor by the nominal openSUSE (that has minimal support). Ergo there is no commercial conflict.

My main PCs typically lag the current openSUSE by a version. BUT I do have two PCs (one sandbox, and one our GNU/Linux Club’s PC) running Tumbleweed. AND I also test every milestone/beta release. What does that make me ?

A lucky and fortunate, motivated man? :slight_smile:

No, I do not share “/home” between the installs. I assume there could be conflict between settings on the two different versions.

I’m not a big multi-media user, so “/home” is not actually all that big. If I did use a lot of multi-media, I would probably put that in a separate partition, with symlinks from home. And that could be easily shared.

The most important of my files in “/home” are kept consistent between systems. I do this with “rsync” over the network. Basically, I have a script that I run on systems that synchronize with my primary desktop system.

I do try to carry out a lot of my day-to-day work using the test install. I don’t try that with tex/latex, though. It’s a big package, so I usually don’t install that on the milestones. And I’m not sure about multi-media on a milestone, because the packman repos are probably not ready for that.

On 2012-01-01 13:46, stakanov wrote:

> The thing about an independent partition interest me, but I would not
> boot often if I cannot share the home. And if I share the home I may run
> into a configuration problem. How do you address this?

I don’t share the home. I mount it aside if needed for data files.

> BTW, although I
> am running a 500 GB disk, incredibly space is never enough. So space is
> also an problem I see with this, but the idea is intriguing me quite
> some time now.

My laptop factory test partition is only 7 GiB. Suffices.

> Although, to be meaningful, testing should be done with
> day by day working practice when the real bugs come out. No?

Yes. But I usually find many more essential bugs before I can try those
things. There are so many problems that I don’t have the opportunity to
test more.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)

On 2012-01-01 05:06, swerdna wrote:

> openSUSE, those packages can then find their way into the enterprise
> level products (SLES and SLED). So it’s really a multi-tier process from
> Factory through the public product openSUSE to the final level, the
> Enterprise level.

Some SUSE employees would deny the relation - which is obvious to me. I
agree with your description.


Cheers / Saludos,

Carlos E. R.
(from 11.4 x86_64 “Celadon” at Telcontar)