[opensuse-factory] 12.2 not suitable to replace soon out of support 11.4
[I started writing this 28 Aug, got interrupted, and just hadn't been able to find time to come back to it until now.] Things that used for work just fine in 11.4 or prior don't any more. Many of these are attributable to systemd replacement of sysvinit for init. Some can be worked around by booting with sysvinit instead of systemd. Some are old bugs, some of which filed upstream and either not filed in bugzilla.novell.com at all or not until late in 12.2's development cycle. KDM kcm module conflicts with /etc/sysconfig/displaymanager settings https://bugzilla.novell.com/show_bug.cgi?id=267903 Filed 65 months ago. :-( kdmrc should be in /etc instead of /usr https://bugzilla.novell.com/show_bug.cgi?id=651897 This is not the only instance of openSUSE FHS violation and frustrating complication to backup/restore processes. Other distros put it in /etc/kd*. Why doesn't openSUSE? KMS broke traditional framebuffer modes https://bugzilla.novell.com/show_bug.cgi?id=675793 This installer bug was filed against 11.4 late, but allowed plenty of time for it to be fixed for 12.1. It's been since unimproved, remaining negative installation polish. Those who need non-default video= have to do it manually, while unneeded vga= modes are just sitting there creating the illusion that they are relevant while video= is not. systemd-init boot puts messages on tty1 after agetty starts https://bugzilla.novell.com/show_bug.cgi?id=725913 Fix is known, but bug owner deferred implementing until release.next. Workaround: install non-default init system sysvinit-init back and forward icons missing in 3.6.x https://bugzilla.novell.com/show_bug.cgi?id=732670 filed in November but much older. I really don't understand this one, as the same Mozilla.org Firefox binary run on any other distro I've tried not only has them, but they are legible. workaround: none known other than use another distro KBD_NUMLOCK="yes" in /etc/sysconfig/keyboard does not work under systemd init https://bugzilla.novell.com/show_bug.cgi?id=746595 filed 11 February as a split from https://bugzilla.novell.com/show_bug.cgi?id=720486 filed 26 September, over 11 months prior to 12.2 release Progress: works on some systems, works sometimes or not at all on others Workaround: install non-default init system sysvinit-init [systemd][3 on cmdline] X session tty locations not predictable (:1 not always on tty8; :0 not always on tty7) https://bugzilla.novell.com/show_bug.cgi?id=768788 workaround: install non-default init system sysvinit-init No panning with intel or nouveau https://bugzilla.novell.com/show_bug.cgi?id=771521 again late filed against 12.2, but filed upstream as https://bugs.freedesktop.org/show_bug.cgi?id=39949 54 weeks ago, not marked fixed, but patch to fix submitted in April workaround: install ATI video (in most cases probably = none) midnight commander no longer opens in current directory https://bugzilla.novell.com/show_bug.cgi?id=775919 filed only recently, but fixed upstream in a minor version increment prior to version freeze workaround: OBS With systemd booting with 3 on cmdline I always see a message "Dependency failed. Aborted stat of Getty on tty1", but haven't looked for a bug about it. It reminds me which systems are running systemd. Most I install sysvinit-init on because systemd still isn't mature enough to be the default on a "polished" distro release. With all the extra time available to eradicate bugs from 12.2, one would think it would show, but I don't see it. To me it looks only marginally better polished than 12.1, which I consider to be the 2nd worst openSUSE release ever. -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Sat, 2012-09-01 at 12:48 -0400, Felix Miata wrote: <snip> The question is not "should" but "can". The fact of the matter is that SUSE has X dollars and y humans in their company, regardless of current openSUSE supported releases. From that x and y, they have to decide how much to give to openSUSE. They do not have unlimited resources and have to think carefully about what they can give to the Project that doesn't drain from their own priorities for their own products. We can argue time and time again that the last supported release version should be given extended support. But that does not change the reality that the availability of resources remains the same. The only real reality is that if we want longer life support for any given version, then we have to step up as a community and do the work. I don't dispute the validity of your concerns, and I don't think many do, nor am I saying I agree with them either. I'm simply neutral on it. I'm pretty sure SUSE doesn't dispute it either. In fact, there was a recent discussion about keeping the 11.4 repos up for longer, even if support for it ceases. Additionally, if I'm not mistaken, Evergreen will support 11.4 as well. But these constant "should be extended longer" arguments I see will always go nowhere if the community doesn't also couple that arugment with "okay, let's band together and make continued support a reality." And on a side note, I have to admit to some distress when looking at the current list of people coming to the openSUSE Summit in three weeks, and barely see anyone from North America who are vocally active in this factory list not registered yet for the Summit. The Summit is an excellent place for folks who have concerns to come together, discuss and collaborate over a three-day weekend and come up with real solutions. I'm not trying to do a shameless plugin of Summit here, but to point out that constant vocalizing of concerns without effort to develop realistic solutions to those concerns and come together in a more collaborative and productive way seems to be quite prevalent. If you want support to be better... get involved. Can't be more simpler than that. Bryen -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On Sat, Sep 1, 2012 at 10:17 AM, Bryen M Yunashko
On Sat, 2012-09-01 at 12:48 -0400, Felix Miata wrote: <snip>
The question is not "should" but "can".
The fact of the matter is that SUSE has X dollars and y humans in their company, regardless of current openSUSE supported releases. From that x and y, they have to decide how much to give to openSUSE. They do not have unlimited resources and have to think carefully about what they can give to the Project that doesn't drain from their own priorities for their own products.
We can argue time and time again that the last supported release version should be given extended support. But that does not change the reality that the availability of resources remains the same. The only real reality is that if we want longer life support for any given version, then we have to step up as a community and do the work.
I don't dispute the validity of your concerns, and I don't think many do, nor am I saying I agree with them either. I'm simply neutral on it. I'm pretty sure SUSE doesn't dispute it either. In fact, there was a recent discussion about keeping the 11.4 repos up for longer, even if support for it ceases. Additionally, if I'm not mistaken, Evergreen will support 11.4 as well.
But these constant "should be extended longer" arguments I see will always go nowhere if the community doesn't also couple that arugment with "okay, let's band together and make continued support a reality."
And on a side note, I have to admit to some distress when looking at the current list of people coming to the openSUSE Summit in three weeks, and barely see anyone from North America who are vocally active in this factory list not registered yet for the Summit. The Summit is an excellent place for folks who have concerns to come together, discuss and collaborate over a three-day weekend and come up with real solutions.
I'm not trying to do a shameless plugin of Summit here, but to point out that constant vocalizing of concerns without effort to develop realistic solutions to those concerns and come together in a more collaborative and productive way seems to be quite prevalent.
If you want support to be better... get involved. Can't be more simpler than that.
Bryen
-- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
I wish I could be at the summit but it's just not financially feasible. I, too, have priorities and limited resources. But on the topic of 11.4 support, I've been very happy with 12.1 and I don't see a need for 11.4 support at all. I switched from 11.4 to 12.1 late in the pre-release process and never looked back. I *am* concerned about the quality of 12.2; RC2 was the first build I could even use for my computational journalism publishers workbench. I'm clearing off a partition on my workstation for the release, but I'm not at all sure I'll ever run it on my laptop on bare metal. As long as I have a solid kernel, compilers, tool chain and LXDE desktop I can run almost anything from upstream source or binaries on *any* distro. 12.2, or for that matter Fedora 17 LXDE or Lubuntu 12.04 LTS aren't a *compelling* improvement over 12.1 for that use case. -- Twitter: http://twitter.com/znmeb; Computational Journalism Publishers Workbench: http://j.mp/QCsXOr How the Hell can the lion sleep with all those people singing "A weem oh way!" at the top of their lungs? -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
El 01/09/12 12:48, Felix Miata escribió:
With systemd booting with 3 on cmdline I always see a message "Dependency failed. Aborted stat of Getty on tty1", but haven't looked for a bug about it. It reminds me which systems are running systemd. Most I install sysvinit-init on because systemd still isn't mature enough to be the default on a "polished" distro release.
None of your reports are bugs in systemd but either somewhere else or systemd just works differently. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Cristian Rodríguez wrote:
El 01/09/12 12:48, Felix Miata escribió:
With systemd booting with 3 on cmdline I always see a message "Dependency failed. Aborted stat of Getty on tty1", but haven't looked for a bug about it. It reminds me which systems are running systemd. Most I install sysvinit-init on because systemd still isn't mature enough to be the default on a "polished" distro release.
None of your reports are bugs in systemd but either somewhere else or systemd just works differently.
Why would you expect people to jump in and file reports immediately when they have the experience of being told they were invalid? OR when they filed bugs 5 years ago that are still open ? (Felix mentioned the kcm bug being 5 years old and unaddressed -- compared to suse's release cycle, that would indicate there was no point in filing that bug). I love it when people say to file bugs -- which I've done, but I've gotten crapped on enough to know that one does so, minimally with wasting one's time, and often being told that your your "XXX" makes things invalid. Some of these, like not supporting product builds if you have related packages installed, I feel are unreasonable, but I'm told in no kind terms that it's my official suse install that is the problem. (if you have a "." on your keyboard, then your bug is invalid...we don't support that..." <- Then people go on about how difficult it is made for people to sign up to file a bug report?!?!?! How long did that thread go on -- They file 1 report and get treatment like I regularly get -- the no signup solution will fix that. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Felix Miata schrieb:
back and forward icons missing in 3.6.x https://bugzilla.novell.com/show_bug.cgi?id=732670 filed in November but much older. I really don't understand this one, as the same Mozilla.org Firefox binary run on any other distro I've tried not only has them, but they are legible. workaround: none known other than use another distro
This should be marked invalid if it doesn't happen with Firefox 14 or 15. 3.6.x hasn't been supported by the upstream for a while and has a growing list of known unpatched security issues. Robert Kaiser -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Probably should have mentioned this installation bug: resolv.conf missing when NetworkManager tabooed http://download.opensuse.org/distribution/12.2/ -- "The wise are known for their understanding, and pleasant words are persuasive." Proverbs 16:21 (New Living Translation) Team OS/2 ** Reg. Linux User #211409 ** a11y rocks! Felix Miata *** http://fm.no-ip.com/ -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (6)
-
Bryen M Yunashko
-
Cristian Rodríguez
-
Felix Miata
-
Linda Walsh
-
M. Edward (Ed) Borasky
-
Robert Kaiser