June 2021 - openSUSE Kernel Bugs - openSUSE Mailing Lists

[Bug 1186710] wrong/dead symlinks with kernel-default-devel-5.12.4-2.1.x86_64 (after usrmerge?)
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186710 https://bugzilla.suse.com/show_bug.cgi?id=1186710#c14 --- Comment #14 from Takashi Iwai <tiwai(a)suse.com> --- (In reply to Michal Suchanek from comment #13) > TW kernel packages have to be built separately for Leap because kernel > contains userspace binaries that require the new libc. For Nvidia or such that needs a build of modules afterwards, yes, but currently kernel-default.rpm itself could be used on Leap until now. > So long as the TW packages work on TW we should be fine for now. > > In the future we might get to a point that we need to rebuild the kernel for > Leap just to fix the symlinks because it would be compatible otherwise. Right, this has to be clarified publicly in wider audience. The incompatibility also includes the all firmware packages and KMPs. It's a challenge how to provide KMPs for Leap with the latest stable kernel, though; we might need to set up OBS Kernel:stable:Backport:KMP for a collection of KMPs to be built with Kernel:stable:Backport. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1183890] [Build 20210320] openQA test fails in welcome - NET ISO is stuck on linuxrc blue screen
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1183890 https://bugzilla.suse.com/show_bug.cgi?id=1183890#c7 Fabian Vogt <fvogt(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |fvogt(a)suse.com --- Comment #7 from Fabian Vogt <fvogt(a)suse.com> --- Same as bug 1184924, but for TW? -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186726] New: XFS: Metadata corruption detected at __xfs_buf_mark_corrupt
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186726 Bug ID: 1186726 Summary: XFS: Metadata corruption detected at __xfs_buf_mark_corrupt Classification: openSUSE Product: openSUSE Distribution Version: Leap 15.2 Hardware: Other OS: Other Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: martin.wilck(a)suse.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- XFS ended up corrupted after the BUG reported in bug 1186724. kernel 5.3.18-lp152.66-default. This system uses ecryptfs home on to of XFS. XFS started logging errors ~25s after the BUG message (full log in https://bugzilla.suse.com/attachment.cgi?id=849866) This happend on April 1, 2021. > [362091.622962] artemis.mittagstun.de kernel: XFS (dm-0): xlog_space_left: head behind tail > [362091.622973] artemis.mittagstun.de kernel: XFS (dm-0): tail_cycle = 2348, tail_bytes = 1957888 > [362091.622978] artemis.mittagstun.de kernel: XFS (dm-0): GH cycle = 2348, GH bytes = 1933664 > [362091.622982] artemis.mittagstun.de kernel: XFS (dm-0): xlog_space_left: head behind tail > [362091.622987] artemis.mittagstun.de kernel: XFS (dm-0): tail_cycle = 2348, tail_bytes = 1957888 > [362091.622991] artemis.mittagstun.de kernel: XFS (dm-0): GH cycle = 2348, GH bytes = 1933664 > [362091.623082] artemis.mittagstun.de kernel: XFS (dm-0): xlog_space_left: head behind tail > [362091.623087] artemis.mittagstun.de kernel: XFS (dm-0): tail_cycle = 2348, tail_bytes = 1957888 > [362091.623092] artemis.mittagstun.de kernel: XFS (dm-0): GH cycle = 2348, GH bytes = 1944348 > [362091.623096] artemis.mittagstun.de kernel: XFS (dm-0): xlog_space_left: head behind tail Eventually the system rebooted. > [ 15.884313] artemis kernel: SGI XFS with ACLs, security attributes, no debug enabled > [ 15.905008] artemis kernel: audit: type=1400 audit(1617303345.346:6): apparmor="STATUS" operation="profile_load" profile="unconfined" name="klogd" pid=1028 comm="apparmor_parser" > [ 15.905832] artemis kernel: XFS (dm-1): Mounting V4 Filesystem > [ 15.989685] artemis kernel: XFS (dm-0): Mounting V4 Filesystem > [ 16.293642] artemis kernel: XFS (dm-1): Starting recovery (logdev: internal) > [ 16.402584] artemis kernel: XFS (dm-1): Ending recovery (logdev: internal) > [ 16.408717] artemis kernel: XFS (dm-0): Starting recovery (logdev: internal) > [ 17.015270] artemis kernel: XFS (dm-0): Ending recovery (logdev: internal) After I logged in and ecryptfs was mounted, I saw the issue > [ 349.629100] artemis.mittagstun.de kernel: XFS (dm-0): Metadata corruption detected at __xfs_buf_mark_corrupt+0xe/0x20 [xfs], xfs_agi block 0x1900010 > [ 349.629102] artemis.mittagstun.de kernel: XFS (dm-0): Unmount and run xfs_repair > [ 349.629156] artemis.mittagstun.de kernel: XFS (dm-0): Internal error xfs_trans_cancel at line 1073 of file ../fs/xfs/xfs_trans.c. Caller xfs_remove+0x19e/0x2d0 [xfs] > [ 349.629160] artemis.mittagstun.de kernel: CPU: 2 PID: 2951 Comm: gvfsd-metadata Not tainted 5.3.18-lp152.66-default #1 openSUSE Leap 15.2 > [ 349.629161] artemis.mittagstun.de kernel: Hardware name: FUJITSU LIFEBOOK S904/FJNB272, BIOS Version 1.20 07/25/2014 > [ 349.629161] artemis.mittagstun.de kernel: Call Trace: > [ 349.629169] artemis.mittagstun.de kernel: dump_stack+0x66/0x8b > [ 349.629222] artemis.mittagstun.de kernel: xfs_trans_cancel+0x12a/0x150 [xfs] > [ 349.629272] artemis.mittagstun.de kernel: xfs_remove+0x19e/0x2d0 [xfs] > [ 349.629320] artemis.mittagstun.de kernel: xfs_vn_unlink+0x55/0xa0 [xfs] > [ 349.629325] artemis.mittagstun.de kernel: ? may_delete+0x79/0x180 > [ 349.629327] artemis.mittagstun.de kernel: vfs_unlink+0x106/0x1f0 > [ 349.629335] artemis.mittagstun.de kernel: ecryptfs_do_unlink+0x7d/0x170 [ecryptfs] > [ 349.629338] artemis.mittagstun.de kernel: vfs_unlink+0x106/0x1f0 > [ 349.629341] artemis.mittagstun.de kernel: do_unlinkat+0x2ba/0x320 > [ 349.629346] artemis.mittagstun.de kernel: do_syscall_64+0x65/0x1f0 > [ 349.629350] artemis.mittagstun.de kernel: entry_SYSCALL_64_after_hwframe+0x44/0xa9 > [ 349.629353] artemis.mittagstun.de kernel: RIP: 0033:0x7f61067f7cf7 > [ 349.629355] artemis.mittagstun.de kernel: Code: f0 ff ff 73 01 c3 48 8b 0d 9e 51 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 66 90 b8 57 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d 71 51 2c 00 f7 d8 64 89 01 48 > [ 349.629357] artemis.mittagstun.de kernel: RSP: 002b:00007ffebcac3d98 EFLAGS: 00000202 ORIG_RAX: 0000000000000057 > [ 349.629359] artemis.mittagstun.de kernel: RAX: ffffffffffffffda RBX: 00005653c129df60 RCX: 00007f61067f7cf7 > [ 349.629361] artemis.mittagstun.de kernel: RDX: 0000000000000028 RSI: 0000000000000000 RDI: 00005653c181bc40 > [ 349.629362] artemis.mittagstun.de kernel: RBP: 00005653c181bc40 R08: 0000000000000000 R09: 0000000000000000 > [ 349.629363] artemis.mittagstun.de kernel: R10: 0000000000000001 R11: 0000000000000202 R12: 00000000682f2784 > [ 349.629364] artemis.mittagstun.de kernel: R13: 00005653c181b400 R14: 00005653c180e320 R15: 00005653c12962a0 > [ 349.629372] artemis.mittagstun.de kernel: XFS (dm-0): xfs_do_force_shutdown(0x8) called from line 1074 of file ../fs/xfs/xfs_trans.c. Return address = 00000000f0fbc3d0 > [ 349.629373] artemis.mittagstun.de kernel: XFS (dm-0): Corruption of in-memory data detected. Shutting down filesystem > [ 349.629374] artemis.mittagstun.de kernel: XFS (dm-0): Please unmount the filesystem and rectify the problem(s) > [ 349.629382] artemis.mittagstun.de kernel: Error in vfs_unlink; rc = [-117] > [ 349.629554] artemis.mittagstun.de kernel: Error opening lower file for lower_dentry [0x000000005209f793] and lower_mnt [0x0000000047a76273]; rc = [-5] > [ 349.629556] artemis.mittagstun.de kernel: ecryptfs_open: Error attempting to initialize the lower file for the dentry with name [root]; rc = [-5] > [ 349.629845] artemis.mittagstun.de kernel: Error opening lower file for lower_dentry [0x000000005209f793] and lower_mnt [0x0000000047a76273]; rc = [-5] > [ 349.629847] artemis.mittagstun.de kernel: ecryptfs_open: Error attempting to initialize the lower file for the dentry with name [root]; rc = [-5] > [ 350.052544] artemis.mittagstun.de unknown[2951]: meta_tree_flush_locked: tree->root == NULL, possible data loss > corrupted file was moved to: /home/martin/.local/share/gvfs-metadata/root.backup.2021-04-01T19:01:19.071895Z > (please make a comment on https://bugzilla.gnome.org/show_bug.cgi?id=598561 and attach the corrupted file) > [ 350.059894] artemis.mittagstun.de gvfsd-metadata[2951]: ** > [ 350.059894] artemis.mittagstun.de gvfsd-metadata[2951]: GVFS:ERROR:../metadata/metatree.c:2419:meta_tree_flush_locked: assertion failed: (res) > [ 350.059894] artemis.mittagstun.de gvfsd-metadata[2951]: Bail out! GVFS:ERROR:../metadata/metatree.c:2419:meta_tree_flush_locked: assertion failed: (res) Unfortunately, I didn't realize this corruption immediately, and thus didn't run xfs_repair at the first opportunity as I should have. The system appeared to work fine after the next reboot. Late issues have made me realize, though, that XFS issues had persisted until the present day. -- You are receiving this mail because: You are the assignee for the bug.

1 5

[Bug 1186710] wrong/dead symlinks with kernel-default-devel-5.12.4-2.1.x86_64 (after usrmerge?)
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186710 https://bugzilla.suse.com/show_bug.cgi?id=1186710#c13 --- Comment #13 from Michal Suchanek <msuchanek(a)suse.com> --- TW kernel packages have to be built separately for Leap because kernel contains userspace binaries that require the new libc. So long as the TW packages work on TW we should be fine for now. In the future we might get to a point that we need to rebuild the kernel for Leap just to fix the symlinks because it would be compatible otherwise. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186363] [Build 20210521] openQA test fails in kdump_and_crash
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186363 https://bugzilla.suse.com/show_bug.cgi?id=1186363#c2 Fabian Vogt <fvogt(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |IN_PROGRESS --- Comment #2 from Fabian Vogt <fvogt(a)suse.com> --- (In reply to Fabian Vogt from comment #1) > The issue is that a preceding module runs "zypper install libvirt vhostmd", > but libvirt -> libvirt-daemon-driver-storage -> nfs-utils -> (kmod(nfsd.ko) > if kernel), but kernel-default-base doesn't have nfsd.ko and so it installs > kernel-kvmsmall instead. > > Due to bad luck, kernel-kvmsmall became the default selection in grub and > was booted instead. Due to good luck, the system came up properly. The > kdump_and_crash test however expects kernel-default(-base) and installs the > wrong -debuginfo package. > > IMO adding nfsd.ko to kernel-default-base and Suggests: kernel-default to > patterns-base-base is the proper solution here. kernel-default-base: https://build.opensuse.org/request/show/896873 patterns-base: https://build.opensuse.org/request/show/896874 -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186729] New: Hibernation stopped working after update 04 May 2020
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

http://bugzilla.opensuse.org/show_bug.cgi?id=1186729 Bug ID: 1186729 Summary: Hibernation stopped working after update 04 May 2020 Classification: openSUSE Product: openSUSE Tumbleweed Version: Current Hardware: x86-64 OS: openSUSE Tumbleweed Status: NEW Severity: Normal Priority: P5 - None Component: Kernel Assignee: kernel-bugs(a)opensuse.org Reporter: rtsvetkov(a)suse.com QA Contact: qa-bugs(a)suse.de Found By: --- Blocker: --- User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:88.0) Gecko/20100101 Firefox/88.0 Build Identifier: Hibernation stopped working after update 04 May 2020. DELL precision 5510. No BIOS updated. perhaps this output from dmesg|grep -i acpi could be useful: [ 15.214389] ACPI BIOS Error (bug): Could not resolve symbol [\_SB.PCI0.LPCB.HEC.ECAV], AE_NOT_FOUND [ 15.214398] ACPI Error: Aborting method \_TZ.TZ01._TMP due to previous error (AE_NOT_FOUND) Reproducible: Always Steps to Reproduce: 1. In gnome choose [suspend] ( it was hibernating to disk) 2. 3. Actual Results: the system doesn't hibernate to disk but by the next start boots a new Expected Results: the system is hibernated to disk -- You are receiving this mail because: You are the assignee for the bug.

1 1

[Bug 1186710] wrong/dead symlinks with kernel-default-devel-5.12.4-2.1.x86_64 (after usrmerge?)
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186710 https://bugzilla.suse.com/show_bug.cgi?id=1186710#c12 --- Comment #12 from Takashi Iwai <tiwai(a)suse.com> --- (In reply to Ludwig Nussel from comment #11) > (In reply to Takashi Iwai from comment #5) > > I'm afraid that the relative symlink is bad in this case. The kernel > > package from the latest build is often used on a Leap or SLE system, and > > this would break things. > > Not sure if that is a good way forward. That limits innovation in the kernel > package. You could still build the same sources for other targets and make > the package behave differently. That's what user space packages usually do > after all, you hardly take the TW binaries :) Yeah, after considering again, I guess there will be also other breakage. After all, TW packages won't be usable any longer on SLE / Leap, unfortunately, even for noarch package like firmware. This has to be clarified somewhere. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186363] [Build 20210521] openQA test fails in kdump_and_crash
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186363 https://bugzilla.suse.com/show_bug.cgi?id=1186363#c1 Fabian Vogt <fvogt(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- CC| |fvogt(a)suse.com --- Comment #1 from Fabian Vogt <fvogt(a)suse.com> --- The issue is that a preceding module runs "zypper install libvirt vhostmd", but libvirt -> libvirt-daemon-driver-storage -> nfs-utils -> (kmod(nfsd.ko) if kernel), but kernel-default-base doesn't have nfsd.ko and so it installs kernel-kvmsmall instead. Due to bad luck, kernel-kvmsmall became the default selection in grub and was booted instead. Due to good luck, the system came up properly. The kdump_and_crash test however expects kernel-default(-base) and installs the wrong -debuginfo package. IMO adding nfsd.ko to kernel-default-base and Suggests: kernel-default to patterns-base-base is the proper solution here. -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186710] wrong/dead symlinks with kernel-default-devel-5.12.4-2.1.x86_64 (after usrmerge?)
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186710 Ludwig Nussel <lnussel(a)suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Blocks| |1029961 -- You are receiving this mail because: You are the assignee for the bug.

1 0

[Bug 1186710] wrong/dead symlinks with kernel-default-devel-5.12.4-2.1.x86_64 (after usrmerge?)
by bugzilla_noreply＠suse.com 02 Jun '21

02 Jun '21

https://bugzilla.suse.com/show_bug.cgi?id=1186710 https://bugzilla.suse.com/show_bug.cgi?id=1186710#c11 --- Comment #11 from Ludwig Nussel <lnussel(a)suse.com> --- (In reply to Takashi Iwai from comment #5) > I'm afraid that the relative symlink is bad in this case. The kernel > package from the latest build is often used on a Leap or SLE system, and > this would break things. Not sure if that is a good way forward. That limits innovation in the kernel package. You could still build the same sources for other targets and make the package behave differently. That's what user space packages usually do after all, you hardly take the TW binaries :) -- You are receiving this mail because: You are the assignee for the bug.

1 0