Announcement

Collapse
No announcement yet.

Fedora 24 Users: Don't Run "DNF Update" From The Desktop

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

  • #41
    Originally posted by AdamW View Post

    Anything you see from Zbyszek in that bug report is just as 'official' as anything you see from Lennart, since Zbyszek is a core systemd maintainer as well. They are discussing the situation there.
    http://i.imgur.com/wj0JGz8.jpg

    Comment


    • #42
      Would this affect anything if DNF was ran through a systemd service, say with something like dnf-automatic?

      Comment


      • #43
        Originally posted by Espionage724 View Post
        Would this affect anything if DNF was ran through a systemd service, say with something like dnf-automatic?
        In that case the dnf process itself would survive OK. If you happen to have X running at the time the dnf run happens, and your hardware is vulnerable to the bug, your X session would still crash (so as the user you'd just kinda be working away and then suddenly, boom, X crash - if you poked through the logs afterwards you'd probably be able to figure out what happened).

        Comment


        • #44
          Originally posted by mbohun View Post
          a little recap:

          1. a problem is reported
          2. all available evidence points (once again obviously) to a bug/problem in systemd
          3. however the pro-systemd hecklers/zealots (in general they are illiterate) unable to tell the diff between tech start the usual lies and mindless denial:



          or

          https://bugzilla.redhat.com/show_bug.cgi?id=1378974#c6


          4. only to admit couple of hours later that yes indeed the problem is with systemd
          https://bugzilla.redhat.com/show_bug.cgi?id=1378974#c8
          [/FONT] 5. a new version of systemd is being released:
          http://i.imgur.com/wj0JGz8.jpg
          look, if you're not going to be constructive, just be quiet.

          We don't know yet for sure precisely what triggers the bug. The systemd-logind thing is a theory that we haven't actually confirmed yet. But yes, if that turns out to be the cause, this would be approximately 50% a systemd bug (and 50% an Xorg bug). I hope that makes you super happy.

          Comment


          • #45
            AdamW Has systemd-229-16.fc24 fixed the bug, so that fresh installations of Fedora 24 are not be affected by this, when they pull an update? Just asking because the bug is closed already. In the comments you just discussed such a fix for fresh installations but it did not sound like systemd-229-16.fc24 will fix it.

            Comment


            • #46
              Originally posted by theghost View Post
              AdamW Has systemd-229-16.fc24 fixed the bug, so that fresh installations of Fedora 24 are not be affected by this, when they pull an update? Just asking because the bug is closed already. In the comments you just discussed such a fix for fresh installations but it did not sound like systemd-229-16.fc24 will fix it.
              Well, it's a bit complicated. 229-16.fc24 fixes the bug, in a sense, by taking the service restart out of the `%postun` scriptlet. If you do a *network* install of F24 now, you should never be affected by this specific bug. But given the nature of the bug, if you do a fresh install from a live image, or DVD, or disk image - then the first time you update the system, going from the original release version of systemd to 229.16-.fc24 - the bug can still happen.

              We're discussing ways to pre-empt the old package's `%postun` in the bug ATM, as a way to try and make sure people don't hit the bug in that scenario.

              Comment


              • #47
                @AdamW,

                Thanks for the update.
                FWIW I've got around ~50 F24 machines, thus far no issues in any of them.
                I assume I'm not getting hit by this bug as I'm running dnf remotely within a screen session?

                - Gilboa
                DEV: Intel S2600C0, 2xE52658V2, 32GB, 4x2TB + 2x3TB, GTX1080, F27/x86_64, Dell UP3216Q 4K.
                SRV: Intel S5520SC, 2xX5680, 36GB, 4x2TB, GTX550, F27/x86_64, Dell U2711..
                BAK: Tyan Tempest i5400XT, 2xE5335, 8GB, 3x1.5TB, 9800GTX, F27/x86-64.
                LAP: ASUS Strix GL502V, i7-6700HQ, 32GB, 1TB+256GB, 1070M, F27/x86_64.

                Comment


                • #48
                  Originally posted by AdamW View Post

                  Well, it's a bit complicated. 229-16.fc24 fixes the bug, in a sense, by taking the service restart out of the `%postun` scriptlet. If you do a *network* install of F24 now, you should never be affected by this specific bug. But given the nature of the bug, if you do a fresh install from a live image, or DVD, or disk image - then the first time you update the system, going from the original release version of systemd to 229.16-.fc24 - the bug can still happen.

                  We're discussing ways to pre-empt the old package's `%postun` in the bug ATM, as a way to try and make sure people don't hit the bug in that scenario.
                  is this Bug in F25 ?

                  Comment

                  Working...
                  X