5 ms·

This supports my initial reply. This was an f2fs bug in a bleeding edge kernel released only seven days earlier. The thread suggests perfectly reasonable mitiga

by b5n 23d ago

This supports my initial reply. This was an f2fs bug in a bleeding edge kernel released only seven days earlier. The thread suggests perfectly reasonable mitigations, especially for an arch user: use the lts kernel or downgrade to 6.8.9. If you're not familiar with linux and/or don't want to deal with trivial issues periodically, don't hang out at the bleeding edge. There are plenty of boring and/or beginner friendly choices out there. I'm not saying linux is perfect, but your conclusions in this instance appear to be uninformed rather than supported by the facts.

1970-01-01 23d ago
Ok fair point, here's plenty more of the same on stable. My point is Linux is always fine right until it isn't. Then you're back to a day of debugging your update. https://bugzilla.kernel.org/show_bug.cgi?id=218770 https://bugzilla.kernel.org/show_bug.cgi?id=218770 https://lore.kernel.org/linux-f2fs-devel/20240409203411.1885121-3-jaegeuk@kernel.org/ https://lore.kernel.org/linux-f2fs-devel/20240409203411.1885... https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07280.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07280.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07260.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07260.html https://bugzilla.suse.com/show_bug.cgi?id=1226043 https://bugzilla.suse.com/show_bug.cgi?id=1226043 https://lists.opensuse.org/archives/list/bugs@lists.opensuse.org/message/JRKH5CFGCTLNXOWUKPVV5OJIH45PUYGI/ https://lists.opensuse.org/archives/list/bugs@lists.opensuse... https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1547844 https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/15478... https://bugs.launchpad.net/bugs/63175 https://bugs.launchpad.net/bugs/63175
- b5n 23d ago
  Most of these links weaken your claim rather than support it. The lkml links are patch submissions in a _development_ discussion, not ordinary users running a stable distro. The f2fs bugzilla links are the same issue we already discussed, a regression in the brand-new 6.9 kernel released only days earlier, with the fix landing in 6.9.3. The suse bug is an upstream e2fsprogs issue. This is precisely the kind of issue distro package maintainers can catch, patch, or avoid before it reaches ordinary users on a stable release. The launchpad link is another development discussion, not an example of ordinary users experiencing a bug in a stable release. Linux does experience regressions, but your argument conflates upstream development, bleeding edge releases, and stable distro releases. Those distinctions are fundamental to how linux distributions work, and these links don't support the claim you're making.
  - 1970-01-01 22d ago
    >Most of these links weaken your claim rather than support it. Yep. Good catch, I think you win the argument. These are all distros that are under development and their underlying issues should never be called-out as handing the user frustration as they willingly chose to become a victim of the kernel.

by b5n 23d ago

1970-01-01 23d ago

Ok fair point, here's plenty more of the same on stable. My point is Linux is always fine right until it isn't. Then you're back to a day of debugging your update. https://bugzilla.kernel.org/show_bug.cgi?id=218770 https://bugzilla.kernel.org/show_bug.cgi?id=218770 https://lore.kernel.org/linux-f2fs-devel/20240409203411.1885121-3-jaegeuk@kernel.org/ https://lore.kernel.org/linux-f2fs-devel/20240409203411.1885... https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07280.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07280.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07260.html https://lkml.iu.edu/hypermail/linux/kernel/2511.2/07260.html https://bugzilla.suse.com/show_bug.cgi?id=1226043 https://bugzilla.suse.com/show_bug.cgi?id=1226043 https://lists.opensuse.org/archives/list/bugs@lists.opensuse.org/message/JRKH5CFGCTLNXOWUKPVV5OJIH45PUYGI/ https://lists.opensuse.org/archives/list/bugs@lists.opensuse... https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/1547844 https://bugs.launchpad.net/ubuntu/+source/systemd/+bug/15478... https://bugs.launchpad.net/bugs/63175 https://bugs.launchpad.net/bugs/63175

b5n 23d ago
Most of these links weaken your claim rather than support it. The lkml links are patch submissions in a _development_ discussion, not ordinary users running a stable distro. The f2fs bugzilla links are the same issue we already discussed, a regression in the brand-new 6.9 kernel released only days earlier, with the fix landing in 6.9.3. The suse bug is an upstream e2fsprogs issue. This is precisely the kind of issue distro package maintainers can catch, patch, or avoid before it reaches ordinary users on a stable release. The launchpad link is another development discussion, not an example of ordinary users experiencing a bug in a stable release. Linux does experience regressions, but your argument conflates upstream development, bleeding edge releases, and stable distro releases. Those distinctions are fundamental to how linux distributions work, and these links don't support the claim you're making.
- 1970-01-01 22d ago
  >Most of these links weaken your claim rather than support it. Yep. Good catch, I think you win the argument. These are all distros that are under development and their underlying issues should never be called-out as handing the user frustration as they willingly chose to become a victim of the kernel.