centos-devel ml feels like /dev/null

Submitted by dag on Thu, 2011/05/05 - 18:53

Today I resigned from the centos-devel mailinglist, after yet another long thread where the core CentOS devs basically ignore the issues raised and use smoke-and-mirror techniques and personal attacks to obfuscate the discussion.

I have been through this recently with the CentOS 5.6 release delays, today it was regarding confusing version changes that make it hard to compare or relate CentOS packages with RHEL packages. After a whole bunch of hypothetical cases and even hard evidence of the past, a mistake I made was sufficient to start the name-calling, I don't know why I even bother. I don't think I deserve this anyway.

So I am out. No need for me to be on the CentOS-devel mailinglist if everything I say is being twisted, mischaracterized or ignored. I can use my free time for more interesting things.

So long CentOS, and thanks for all the fish.

Update: Since there seems to be some confusion about my involvement in CentOS. I had left the CentOS core team two years before announcing that I left the centos-devel mailinglist.

It it sad to see centos going

It it sad to see centos going down the drain. Any other RHEL 'lookalike' you can recommend instead?

Scientific Linux

I'd go with Scientific Linux. Their process is more open and the people are actually friendly to feedback.

Are you going to support SL?

Hi Dag,

I had a look to Scientific Linux 6 a couple of months ago, boring of waiting CentOS6 to be released and looked a very interesting alternative.

My question is: Will you support Scientific Linux packaging rpms for them in your repos?

IMHO that should be possitive for everyone: SL, you and of course the users.

Goog luck with your decission!

Hi, The RPMforge and ELRepo

Hi,

The RPMforge and ELRepo RPM packages have always been supported for RHEL, CentOS _and_ Scientific Linux. It has been like this for years and we don't plan to change this.

What's the root cause of that

What's the root cause of that ? I didn't follow the thread.... Are you stopping your rpm activity ?

No, it is not related to the

No, it is not related to the RPMforge or ELRepo projects.

I don't know why the CentOS project prefers to keep everything closed and doesn't acknowledge that 3 months (CentOS 5.6) and 6 months (CentOS 6.0) is a long time for releasing their rebuilds.

Because the process is pretty much closed, and there's an omerta for everyone involved in the release process, there's little opportunity to see the situation improve.

And by not opening up the process or share experiences, the project doesn't enable other people to experiment or start a competing effort. And despite the name, it's neither a community effort, or Enterprise class. (Enterprises don't wait 3 months for security updates)

It's sad, so sad :-/ And despite all this, I felt responsible for promoting CentOS to people for 4 years.

:(

Dag,

I completely understand how you feel, similar things have happened to me in the professional world.

CentOS owes allot to you to be where it is today, and KB is obviously incapable of leading it properly (no need to respond to that publicly).

SL is a good start, but could use more developers. Perhaps you'll find a "happier" hobby fit there?

Wish you the best with whatever you choose to do in the open source world, or work of RHEL clones.

I for one appreciate all your hardwork with RPMForge and ELRepo, you have saved me and many coworkers time and time again.

-- An Avid Follower
S Crothers

agreed it is a shame to see

agreed it is a shame to see Centos go down the tubes, but your contribution to it, and Linux in general has been well and truely noted.

I saw you left the CentOS

I saw you left the CentOS project, so I upgraded from CentOS 5.6 to Scientific Linux 6. For anyone else wanting to do this, boot with the following grub option from the installer:
upgradeany
Then SL6 will let you upgrade CentOS.

history repeating itself

Dag, I respect your thoughts and attitude towards the Centos team, but to people who are quickly shouting attotv about Scientific Linux, i have a few words to remind that we all face the possibility of history repeating itself in this case. Remember how WhiteBox Linux and other similar efforts fail and how Centos rose to the tide? I admit that things are gloomy in terms of co-operation at centos side, but won't the same issues arise at Scientific Linux in the future? I assume a few ddisgruntled dev will leap over from Centos to SL and the question is how open is SL's release process? and how robust is it?
My sixth sense tells me it would be better to fight on the Centos release process, demand some real changes, demand openess and a clear operating code. There is still a bright future for centos if the team is willing to work things out. The worst would be for all of us to pack and convert to SL. This will set a bad precedence and future efforts on recompiled distributions will will suffer similar fate. This is just my 2 cents, somebody has to fight on for the good of centos. There are a few other guys like Dag who are able to have a voice in the centos community, guys dont let us down...

Well, it's obvious from the

Well, it's obvious from the many threads the past few months that there is great resistance to transparency. The past weeks it has been explicitly stated by Johnny that the 'C' in CentOS only means a 'community' of users can use CentOS freely, but the development will stay a 'closed' process. They don't feel any need or obligation to open the process and make it a proper community project.

That CentOS will not be Enterprise-ready is obvious to me since they don't see a need to improve the process either or make faster releases. They apparently don't mind that a release can be made 3 months after RHEL. And that CentOS users have no security updates for 50% of the time (6 months). So the only part of the name that has any meaning is 'OS'.

To me CentOS has been effectively reduced to being an 'OS' :-/ I give up caring about this, other opportunities or alternatives deserve the effort of the community more than 'OS'...

Re: history repeating itself

Seeing as CERN and Fermilab are the ones putting out Scientific Linux, I don't see it going the way CentOS seems to be going at all.

What you said was quite well

What you said was quite well put together. I personally find the attitude of CentOS's closed development is just unacceptable for a community based project. For a community based project to succeed, there has to be some form of transparency so anyone can see what they're doing.

As of today, we haven't seen any "hey, this is what we're doing right now" release of EL6 from them at the moment. The staff over at CERN/Fermi Accelerator did exhibit this during the development lifecycle, by releasing alpha and beta versions of SL6.

No apologies needed.

"I felt responsible for promoting CentOS to people for 4 years."
Hey, no need to apologise for that..
Despite any issues now, I used You and CentOS as myspringboard to Linux.
My thanks to you.

Sory it has worked out so poorly for you.
Regards
Longboard

One side story is never the whole truth

I read all threads in last 3 months.

Dag, the last name calling was because you accidently send 2 e-mails to dev mailing list where you wrote harsh accusations about CentOS dev team, and then said that you say those harsh words only behind their backs (in private e-mails/communication).

Also, you constanly ignore the fact that there were 4.9 and 5.6 releases before 6.0 and the fact that SL has not released SL5.6 yet. There is no way to objectively say that SL devs have released all 3 releses before CentOS devs and not strech the truth to the braking point.

All of your other debates I can not and will not comment, but my impression is that you have (also) contributed to this prolonged incident.

Release of CentOS 6.0 is scheduled for end of May, ~ two weeks from now, it's in QA now, so there is no need for requiems.

Ljubomir

Ljubomir,There were not

Ljubomir,

There were no harsh accusations, I simply stated in what I thought was a private mail that the CentOS developers have been ignoring the real issues. Nothing new from what I wrote to the mailing list at least two times before. The fact this is represented wrongly, and regardless of the fact that I apologized for sending that mail to the list while it was meant privately, does not make any difference to what I wrote on my blog and the decision I have made.

The discussion is not about CentOS vs Scientific Linux either. Scientific Linux obviously has a more open process and does release everything they have built in Alpha, Beta and RC releases. And they have rolling releases to subscribe to. But even if Scientific Linux would be comparable, I don't think it's a very convincing argument, it doesn't make CentOS any less late, doesn't help for the future and is very unprofessional.

Besides, nowhere did the CentOS team acknowledge that releasing 3 months late (CentOS 5.6) or +6 months late (CentOS 6.0) might be just a tad of a problem. No instead they clearly stated that there is no issue, we can be late if we want to and it's free so you cannot complain.

For this specific thread, it is *very* clear that if they simply acknowledged the problem, like they did at the very end when they were put in front of hard evidence, the thread would be 4 posts long, but as usual there's a cover-up and a ridicule phase in everything brought up on the lists. It is a real shame.

With all do respect, you are

With all do respect, you are as bad as they are when it comes to exeggaration. I do not want to get into who started first, or who uses it more. My view from the distance is that you and CentOS devs have personal issues bordering on physical confrontation. I have seen and am seeing both sides twisting the facts to serve their needs. And the truth is you both have some valid points, but as long as (all of) you keep saying that your way is the best way there can be no resolution.

When I said on one of the lists that even my personal life is held up by C6 release, I got personal e-mail with appologies that it takes so long. So devs are not hostile to everyone, but I can feel with (not justify) them when they respond harshly of what they see as personal attacks.

And other part of the problem is that guys on your side are trying to out out a flame with gasoline ("funny way to spell water"), so devs respond in general, and not specificaly to your question/concern. And this happend on both sides and constantly.

As for opennes, term "Open source" does not mean "open process", and they are not obligated to divulge anything, that part is common practice and a norm for development projects. CentOS is a rebuild not a development project, as I see it.

You are correct as to them

You are correct as to them not being hostile to everyone. They are usually only hostile to people that raise a critical voice. Whether this is about release dates, opening up the process or even release tags of RPM packages (which is the thread I was referring to).

And it is very hard, to impossible, to raise a critical voice without one of the core developers taking it personally, instead of rationally. There seems to be a state of us against them.

BTW there are no "guys at my side" and there is no conspiracy. I think that's where the problem lies. But hey, I left centos-devel mailing list and am about to leave the centos mailing list too, so problem solved !

I never meant is as a

I never meant is as a conspiracy. But like it or not, you were a part/side of a full blown flame war that still rages on CentOS mailing list. And all wars have sides. More then once someone "on your side" interjected and replied to e-mail meant for you and made things even worse. And you did too. CentOS devs did the same for each other. Even I sometimes replied with passion. It is a normal behavior for members of current civilization that have same objective, there is nothing extraordinary in that.

I am sure both sides (in this flame war) are at least a little offended by my possition and calling your (all of you involved) name in front of a class, but my hope is that it will be at the end seen as benigne intervention meant to make you all stop and look at your behavior and hopefully calm the situation down.

Ljubomir, I have been using

Ljubomir, I have been using CentOS for some years, from 3.x forward.

I have participated in the mailing lists, but frankly, I felt unwelcome, particularly whenever I ask I asked about plans for expected releases, when I found responses particularly offensive.

I have seen your comments regarding Dag in several places, and I feel they say more about Ljubomir than they do about Dag.

When I speak about someone publicly, I try to be tactful so as to not cause offense. If I were discussing, let us say, your behaviour with someone else, perhaps a friend, I'd not feel the need and I might well use terms which might lift your eyebrows.

I have tried to rebuild RHEL on my own account two or three times. I know it's not trivial, but perhaps if CentOS had documentation on its site. centos.org, about how the build process is done, "here are the tools we are using," and "There is the progress and list of work left to be done," just maybe, I'd have had a look to see what I might do. I've recenly retired, so time isn't a problem, and I have a nice new Quad-core HP Workstation with free disk space.

I'm tossing up between Debian and SL6. It was going to be CentOS 6, but like other folk, I find the delay too discouraging.

Weird way to show gratitude

Cheer up, they'll soon figure out what they have lost. I've read dev-mail for few months and I can say you made right decision.

we should *all* stop reading the mailing-lists ...

Frankly, I stopped reading the mailing-lists (or such long-and-without-end threads) a *long* time ago. This is just time consuming and the time required to just read (and then answer) is precious time that can't be spent on the project itself. So i prefer now working 'backstage' and this is more efficient :-)

Arrfab, Please tell other

Arrfab,

Please tell other CentOS devs to do the same. :-)

OMFG

All you have to do is read something like this from Johnny...

http://lists.centos.org/pipermail/centos/2011-May/111670.html

to know where he is coming from.

The CentOS developers have long since lost any joy in their project and now just see everyone as 'us' or 'them' and 'them' are 'ungrateful bastards'

And to think that Johnny has always been the most reasonable and accessible team member.

I followed your link and

I followed your link and found this:

http://lists.centos.org/pipermail/centos/2011-May/111688.html

I can understand why some developers get frustrated when someone posts something like that.

Moved away from CentOS

Dag, you are the man and you did the right thing. Your contributions are second to none to the enterprise linux community. Closed entOS sooner or later will die. I moved a couple of workstations to Debian 6 and servers to SL 6 so I don't have to deal with Closed entOS again.

Dag I wonder why you waited

Dag I wonder why you waited this long:-)

I read the mailinglist from time to time and I couldn't understand why CentOS Dev don't see a problem with the current speed of development. Their response to critics has been rude and ungrateful for a long time.

This behaviour ,especially the attitude towards a release shedule, caused me a few headaches.

Due to the delayed release of centos 6 I'm currently in trouble not beeing able to provide some long promised features to our devs

Therefore I have been evaluating SL for the past few weeks and we will soon switch to SL to avoid this in the future.

Anyways, keep up the good work!

More bad news for CentOS!

Your voice in CentOS development will be missed by those of us sticking through the current round of issues. There is at least some hope since you still build for SL and CentOS. You are an icon in this field! Thank you for all of your hard work and dedication. I stand on your giant shoulders.

I'm with Dag

I also fled CentOS after this most recent flub (lack of any updates for months at a time). I'm also using SL and feel relief that Dag seems to have chosen it as the "weapon of choice" as well. I'm currently in the midst of gradually migrating a few hundred prod boxes from CentOS to RHEL6 (in a very few cases) and SL6. It's been such a mind bender to send email to SL developers and get a prompt/courteous reply instead of the "if you don't like it, leave" that I'd gotten from more than one dev on the CentOS mailing list (and often no response at all to direct email). Things are looking up in the world.

Best,

Thanks for all your work. And how does SL compare?

Dag,

Thanks very much for all your work over the years, both CentOS specific and outside it. If you didn't exist they would have had to invent you :).

If anyone has good pointers about comparisons of CentOS vs. the other alternatives for RHEL in terms of binary compatibility then I'd appreciate to hear about them. (I see everyone and his fish mention SL above, I just don't want to discard other alternatives).

I run the network of a start-up company which relies heavily on CentOS (I switched from Debian as soon as I learned that it's much easier to find hosting providers which support CentOS when we started) and as much as I'd like to give CentOS credit for their hard work and stable output, a closed system which is even too closed to hear the voices from outside telling them that they are in trouble doesn't look like a solid platform to stay on...

Good move!

I just realized I can't find an official CentOS fix for CVE-2011-3192. Definitely time to move to Scientific Linux.

Oh.. and I did search CentOS' mailing list and forum, look at what I found..

http://lists.centos.org/pipermail/centos/2011-September/117004.html

Good move, Dag!