Announcement

Collapse
No announcement yet.

Debian To Maintain 6.0 Squeeze As An LTS Release

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Debian To Maintain 6.0 Squeeze As An LTS Release

    Phoronix: Debian To Maintain 6.0 Squeeze As An LTS Release

    Regular security support for Debian 6.0 "Squeeze" was set to end after next month, but now the Debian developers have decided to provide security support for this 2011 Debian Linux release until February 2016 -- marking five years since the original Debian 6.0 release...

    http://www.phoronix.com/vr.php?view=MTY2NzA

  • #2
    LTS pointless most of time

    In my opinion a long term linux release is only suitable for servers (and in some cases it isn't), for end users is pointless, you have to deal with outdated software. An example is video drivers, they are still evolving and theres no easy way to install new versions without compiling the whole xserver and linux kernel (something that not everybody is willing to do). Another example would be PHP, that in this case it is version 5.3 on squeeze while 5.4 - 5.5 have lots of optimizations (opcache) that would benefit a server performance. Also newer kernels have introduced power usage improvements making a server be more energy efficient.

    Windows XP has survived a lot because it is easy to download the installer of latest software versions and have it working on your system in a few steps, while on a linux distro you mostly depend on the main software repositories (outdated) or third party repositories which are a pain to deal when upgrading the whole system.

    This LTS releases should adopt a semi rolling release model where new software versions is stability/compatibility tested and rolled out to main repositories when testing is done. Missing the advantages and improvements of newer software versions is a drawback in my opinion. I think LTS is a waste of human resources.

    Comment


    • #3
      Originally posted by TheOne View Post
      In my opinion a long term linux release is only suitable for servers (and in some cases it isn't), for end users is pointless, you have to deal with outdated software. An example is video drivers, they are still evolving and theres no easy way to install new versions without compiling the whole xserver and linux kernel (something that not everybody is willing to do). Another example would be PHP, that in this case it is version 5.3 on squeeze while 5.4 - 5.5 have lots of optimizations (opcache) that would benefit a server performance. Also newer kernels have introduced power usage improvements making a server be more energy efficient.

      Windows XP has survived a lot because it is easy to download the installer of latest software versions and have it working on your system in a few steps, while on a linux distro you mostly depend on the main software repositories (outdated) or third party repositories which are a pain to deal when upgrading the whole system.

      This LTS releases should adopt a semi rolling release model where new software versions is stability/compatibility tested and rolled out to main repositories when testing is done. Missing the advantages and improvements of newer software versions is a drawback in my opinion. I think LTS is a waste of human resources.
      As I understand it, the catalystic event that launched the discussion which eventually resulted in this LTS initiative was when DreamHost announced that they were leaving Debian for Ubuntu LTS because Debian's support lifecycles were too short for them and their tens of thousands of servers. If my understanding is correct, that would appear to confirm your assertion that this is mostly for servers. However, I am not completely sure if I follow your logic that leads you to claiming that such a release is a waste of human resources. Can you expand?

      Comment


      • #4
        Originally posted by Serge View Post
        As I understand it, the catalystic event that launched the discussion which eventually resulted in this LTS initiative was when DreamHost announced that they were leaving Debian for Ubuntu LTS because Debian's support lifecycles were too short for them and their tens of thousands of servers. If my understanding is correct, that would appear to confirm your assertion that this is mostly for servers. However, I am not completely sure if I follow your logic that leads you to claiming that such a release is a waste of human resources. Can you expand?
        Bah, fucking 5 min edit limit. So stupid. Wanted to add the following:

        The reason why I am questioning your logic is because I feel like these kind of companies are more interested in something that works now to continue to work the same + with security protection against newly discovered vulnerabilities and threats than they are in being able to benefit from new optimizations and performance enhancements. When DreamHost announced they were switching, they described how the last Debian migration took them nearly a year to complete across all of their servers, and that this is not economically sustainable for them with two-year distro lifecycles. An organization like that will prioritize multi-year long-term stability, and will not be interested in enhancements that can compromise such stability.

        Comment


        • #5
          It is not wasting of human (Debian) resourses , because it is not handled only by the Debian security team .

          squeeze-lts is not handled by the Debian security team,
          but by a separate group of volunteers and companies interested in
          making it a success
          (with some overlap in people involved). So, if
          you're a company using Debian and seeing a benefit in security
          support for five years, get in touch with team@security.debian.org
          and we'll see how you can help (if you e.g. don't have the manpower /
          know how but are willing to contribute, we can point you to a list
          of Debian consultants)
          * It needs to be pointed out that for this effort to be sustainable
          actual contributions by interested parties are required. squeeze-lts
          is not something that will magically fall from the sky. If you're
          dependent/interested in extended security support you should make an
          effort to contribute, either by contributing on your own or by
          paying a Debian developer/consultant to contribute for you.
          The security team itself is driving the effort, NOT doing it. Some team
          members will contribute to it individually, however.

          Comment


          • #6
            Originally posted by Serge View Post
            However, I am not completely sure if I follow your logic that leads you to claiming that such a release is a waste of human resources. Can you expand?
            Instead of focusing on packaging newer-software, missing-software, fixing bugs, as improving the whole system experience, package developers have to back-port security patches to support companies running old distro versions with old software, companies which may not even donate to the project to support it. Some of this packagers donate their free time, so instead of shaping the future of the OS they are shaping the past, which I see as a waste of time.

            Maybe a semi rolling release model would end all that lost time supporting older releases as doing new distro releases. Also end users would not have to spend so much time doing full system upgrades as salivating for new features on newer software version which they only can obtain by compiling or doing a full system upgrade.

            Comment


            • #7
              The large number of xp uses show that this is also something decided for the desktop

              Comment


              • #8
                Originally posted by dungeon View Post
                It is not wasting of human (Debian) resourses , because it is not handled only by the Debian security team .
                ahh thats some interesting information

                Comment


                • #9
                  Yep and people better to be called that 3+2 support instead of 5, because normal supporting period is on 3 years and will ended 31st May and those +2 can be called extended support for servers .

                  Of course that can be good for workstations too, but of course it is not for desktop . But we will never know for what it can be, because people uses Windows XP even today for their desktops, does not metter it was entered extending phase in year 2009. .

                  Comment


                  • #10
                    Originally posted by dungeon View Post
                    But we will never know for what it can be, because people uses Windows XP even today for their desktops, does not metter it was entered extending phase in year 2009. .
                    I think the fact people still uses XP is because upgrading software is easier, since you download a new installer and you will instantly have access to the latest version of your preferred applications, while on a linux distro you have to compile or depend on packagers, which makes it less attractive in some scenarios.

                    Also still today most windows xp video drivers are more mature that linux ones and you dont have this problem that each time a new kernel/xserver is released drivers are broken.

                    So in other words people can still run windows XP most of the time with latest versions of favorite open source applications with little effort, with optimized drivers (unless they buy new hardware).

                    Comment


                    • #11
                      One can buy new AMD AM1 hardware XP is supported, but it is not supported by Microsoft anymore . So it is not even legacy. it is not supported by company who made it .

                      As of installers, you can use original installers for Firefox, Blender, etc. they work similar to Windows. There are some Linux distributions who can make it that ease way, or games can be packaged for click and run way, for example:

                      http://www.portablelinuxgames.org/


                      Originally posted by TheOne View Post
                      So in other words people can still run windows XP most of the time with latest versions of favorite open source applications with little effort, with optimized drivers (unless they buy new hardware).
                      I can run Debian Sarge with Xfree86 server and fglrx on my Athlon 2200 and Radeon 9250, RTCW will work but laggy as fglrx is ever been . Support ended just like for Window XP, but who cares.. so if do it, that will just for the fun or if i feel boring . So maybe i will try unsupported Windows XP on Athlon 5350 - AMD actually advertised that it is supported, which is cool of course but i don't care
                      Last edited by dungeon; 04-18-2014, 12:36 PM.

                      Comment


                      • #12
                        With x86/AMD64 limited support, it is clearly focused on servers. Also, with RHEL using similar packages versions (think kernel and some other low-level packages), it's not as time-consuming as maintain only by volunteers.

                        I'm not sure how loud were entities demanding extended support for squeeze, as upgrading to wheezy is really easy, if you don't mess with packages version and custom repos. Upping to Jessie will be a real challenge with systemd kicking in (and his various regressions, like journald compared to good old rsyslog).

                        Comment


                        • #13
                          This was obvious, since the International Space Station use Debian 6.

                          Comment


                          • #14
                            Originally posted by TheOne View Post
                            Instead of focusing on packaging newer-software, missing-software, fixing bugs, as improving the whole system experience, package developers have to back-port security patches to support companies running old distro versions with old software, companies which may not even donate to the project to support it. Some of this packagers donate their free time, so instead of shaping the future of the OS they are shaping the past, which I see as a waste of time.

                            Maybe a semi rolling release model would end all that lost time supporting older releases as doing new distro releases. Also end users would not have to spend so much time doing full system upgrades as salivating for new features on newer software version which they only can obtain by compiling or doing a full system upgrade.
                            That's a good point. Backporting efforts do not typically help in terms of overall technological progress (at least not in a direct sense, anyway). However, not everything should be done in the name of forward technological advancement. Effort should be invested in making things practically useful as well, or else the effort invested in research and theory is just as wasteful.

                            Backporting security fixes so that users who cannot for various reasons keep up with the newest releases of software they use are enabled to make use of stable versions of software for longer periods, in my opinion, qualifies as an example of effort invested in making things practically useful. If we are to believe DreamHost's explanation of why they moved to Ubuntu LTS, it is impossible for them to run their business profitably if they have to replace the operating system on all of the servers they maintain every two years. In that case, it is not a question of DreamHost being greedy and unwilling to invest the money necessary to do their part as good open source citizens, as apparently that level of investment would make their business impossible to sustain. That they are able to build a business around using open source software is, in my opinion, still contributing to overall technological progress, just not directly.
                            Last edited by Serge; 04-18-2014, 04:59 PM.

                            Comment


                            • #15
                              So in other words people can still run windows XP most of the time with latest versions of favorite open source applications with little effort, with optimized drivers (unless they buy new hardware).
                              I am a big critic of the linux graphics stack and yet it is better in many ways to the pre-vista stack in Windows. The responsiveness of the stacking WM in Windows in awful. The reason people use XP is the applications. The applications on Windows are always superior, even the ones that have Linux versions.

                              Comment

                              Working...
                              X