Solution disable fsck for rootfs (/).
(In reply to Perfect Gentleman from comment #0) > Solution disable fsck for rootfs (/). Solution is disable fsck for rootfs (/).
rc.log of unsuccessful boot --------------------------- rc default logging started at Sat Nov 24 11:29:13 2018 razerd | * Starting razerd ... rpc.pipefs | * Setting up RPC pipefs ... chronyd | * Starting chronyd ... fsck | * Checking local filesystems ... [ ok ] fsck |Info: Fix the reported corruption. fsck |Info: Mounted device! fsck |Info: Check FS only due to RO fsck |Error: Failed to open the device! fsck | * Filesystems couldn't be fixed [ !! ] fsck | * ERROR: fsck failed to start [ ok ] [ ok ] urandom | * ERROR: cannot start urandom as fsck would not start netmount | * ERROR: cannot start netmount as fsck would not start cronie | * ERROR: cannot start cronie as fsck would not start localmount | * ERROR: cannot start localmount as fsck would not start mtab | * ERROR: cannot start mtab as fsck would not start bootmisc | * ERROR: cannot start bootmisc as fsck would not start elogind | * ERROR: cannot start elogind as fsck would not start net.eno1 | * ERROR: cannot start net.eno1 as fsck would not start dbus | * ERROR: cannot start dbus as fsck would not start root | * ERROR: cannot start root as fsck would not start cupsd | * ERROR: cannot start cupsd as fsck would not start rpcbind | * Starting rpcbind ... [ ok ] rpc.idmapd | * Starting idmapd ... rpc.statd | * Starting NFS statd ... [ ok ] [ ok ] nfsclient | * Starting NFS sm-notify ... [ ok ] nfs | * Mounting nfsd filesystem in /proc ... [ ok ] nfs | * Exporting NFS directories ... nfs |exportfs: could not open /var/lib/nfs/.etab.lock for locking: errno 30 (Read-only file system) nfs |exportfs: can't lock /var/lib/nfs/etab for writing [ !! ] nfs | * Starting NFS mountd ... [ ok ] nfs | * Starting NFS daemon ... [ ok ] nfs | * Starting NFS smnotify ... [ ok ] local | * Starting local ... [ ok ] rc default logging stopped at Sat Nov 24 11:29:13 2018 ----------------------------------------------------
Your divece was mounted, but failed to open, which says nothing special but possible problems in your system settings.
(In reply to Mikle Kolyada from comment #3) > Your divece was mounted, but failed to open, which says nothing special but > possible problems in your system settings. wat? It works with 1.11.0-r1.
rc boot logging started at Sat Nov 24 11:24:22 2018 modules | * Loading module vboxdrv ... [ ok ] modules | * Loading module vboxnetflt ... [ ok ] modules | * Loading module vboxnetadp ... [ ok ] modules | * Loading module vboxpci ... [ ok ] hwclock | * Setting system clock using the hardware clock [UTC] ... [ ok ] sysctl | * Configuring kernel parameters ... hostname | * Setting hostname to HOSTNAME ... loopback | * Bringing up network interface lo ... [ ok ] fsck | * Checking local filesystems ... [ ok ] [ ok ] fsck |Info: Fix the reported corruption. fsck |Info: Mounted device! fsck |Info: Check FS only due to RO fsck |Info: [/dev/sdd1] Disk Model: PLEXTOR PX-G256M1.06 fsck |Info: Segments per section = 1 fsck |Info: Sections per zone = 1 fsck |Info: sector size = 512 fsck |Info: total sectors = 500116111 (244197 MB) fsck |Info: MKFS version fsck | "Linux version 4.15.0-20-generic (buildd@lgw01-amd64-039) (gcc version 7.3.0 (Ubuntu 7.3.0-16ubuntu3)) #21-Ubuntu SMP Tue Apr 24 06:16:15 UTC 2018" fsck |Info: FSCK version fsck | from "Linux version 4.18.0-11-generic (buildd@lcy01-amd64-027) (gcc version 8.2.0 (Ubuntu 8.2.0-7ubuntu1)) #12-Ubuntu SMP Tue Oct 23 19:22:37 UTC 2018" fsck | to "Linux version 4.19.4-gentoo (root@HOSTNAME) (gcc version 8.2.0 (Gentoo 8.2.0-r4 p1.5)) #1 ZEN SMP PREEMPT Fri Nov 23 22:15:59 +07 2018" fsck |Info: superblock features = 0 : fsck |Info: superblock encrypt level = 0, salt = 00000000000000000000000000000000 fsck |Info: total FS sectors = 500116104 (244197 MB) fsck |Info: CKPT version = 7965081c fsck |Info: Checked valid nat_bits in checkpoint fsck |Info: checkpoint state = c5 : nat_bits crc compacted_summary unmount fsck | fsck |[FSCK] Unreachable nat entries [Ok..] [0x0] fsck |[FSCK] SIT valid block bitmap checking [Ok..] fsck |[FSCK] Hard link checking for regular file [Ok..] [0xe0] fsck |[FSCK] valid_block_count matching with CP [Ok..] [0x6bb835] fsck |[FSCK] valid_node_count matcing with CP (de lookup) [Ok..] [0xebfb3] fsck |[FSCK] valid_node_count matcing with CP (nat lookup) [Ok..] [0xebfb3] fsck |[FSCK] valid_inode_count matched with CP [Ok..] [0xeb3ea] fsck |[FSCK] free segment_count matched with CP [Ok..] [0x1a345] fsck |[FSCK] next block offset is free [Ok..] fsck |[FSCK] fixing SIT types fsck |[FSCK] other corrupted bugs [Ok..] fsck | fsck |Done. fsck |Info: Fix the reported corruption. fsck |Info: [/dev/sda1] Disk Model: SAMSUNG MZ7WD2407W3Q fsck |Info: Segments per section = 1 fsck |Info: Sections per zone = 1 fsck |Info: sector size = 512 fsck |Info: total sectors = 468860047 (228935 MB) fsck |Info: MKFS version fsck | "Linux version 4.16.6-gentoo (root@kubuntu) (gcc version 7.3.0 (Gentoo 7.3.0-r1 p1.1)) #1 ZEN SMP PREEMPT Tue May 1 18:28:05 +07 2018" fsck |Info: FSCK version fsck | from "Linux version 4.19.4-gentoo (root@HOSTNAME) (gcc version 8.2.0 (Gentoo 8.2.0-r4 p1.5)) #1 ZEN SMP PREEMPT Fri Nov 23 22:15:59 +07 2018" fsck | to "Linux version 4.19.4-gentoo (root@HOSTNAME) (gcc version 8.2.0 (Gentoo 8.2.0-r4 p1.5)) #1 ZEN SMP PREEMPT Fri Nov 23 22:15:59 +07 2018" fsck |Info: superblock features = 0 : fsck |Info: superblock encrypt level = 0, salt = 00000000000000000000000000000000 fsck |Info: total FS sectors = 468860040 (228935 MB) fsck |Info: CKPT version = 40e5095e fsck |Info: checkpoint state = c5 : nat_bits crc compacted_summary unmount fsck |Info: No error was reported [ ok ] root | * Remounting root filesystem read/write ... [ ok ] root | * Remounting filesystems ... [ ok ] mtab | * Updating /etc/mtab ... swap | * Activating swap devices ... mtab | * Creating mtab symbolic link [ ok ] [ ok ] localmount | * Mounting local filesystems ... [ ok ] urandom | * Initializing random number generator ... [ ok ] bootmisc | * Creating user login records ... [ ok ] termencoding | * Setting terminal encoding [UTF-8] ... dbus | * Starting dbus ... [ ok ] [ ok ] keymaps | * Setting keyboard mode [UTF-8] ... [ ok ] keymaps | * Loading key mappings [us] ... [ ok ] elogind | * Starting elogind ... [ ok ] rc boot logging stopped at Sat Nov 24 11:24:42 2018
I didn't hear any explanation why it works with 1.11.0 but not with 1.12.0
How is that gentoo specific? You did not provide any info, /etc/init.d/fsck fails as it can't run f2fs' fsck command. As per your output fsck mounted your device, but could not open it at least twice. So nothing for gentoo here.
(In reply to Mikle Kolyada from comment #7) > How is that gentoo specific? > > You did not provide any info, /etc/init.d/fsck fails as it can't run f2fs' > fsck command. As per your output fsck mounted your device, but could not > open it at least twice. So nothing for gentoo here. what info else should I provide? there was no changes except f2fs-tools update. if it's not gentoo specific, it must be reported to upstream.
I'm also experiencing the issue and downgrading from sys-fs/f2fs-tools-1.12.0 to f2fs-tools-1.11.0-r1 resolved the issue. I'm happy to provide more information, but I need help with getting the boot information from a systemd system.
Again, go to upstream please.
Reopening as reporter contacted upstream [1] and a working patch is available (see URL). [1] https://sourceforge.net/p/linux-f2fs/mailman/message/36476197/
*** Bug 671866 has been marked as a duplicate of this bug. ***
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=76ae45b316e09ec982a1f02d77017efbd5909d40 commit 76ae45b316e09ec982a1f02d77017efbd5909d40 Author: Mikle Kolyada <zlogene@gentoo.org> AuthorDate: 2018-12-03 14:12:19 +0000 Commit: Mikle Kolyada <zlogene@gentoo.org> CommitDate: 2018-12-03 14:12:41 +0000 sys-fs/f2fs-tools: revbump to fix fsck on boot Closes: https://bugs.gentoo.org/671786 Signed-off-by: Mikle Kolyada <zlogene@gentoo.org> Package-Manager: Portage-2.3.51, Repoman-2.3.11 sys-fs/f2fs-tools/f2fs-tools-1.12.0-r1.ebuild | 37 ++++++++++++++ .../f2fs-tools/files/f2fs-tools-1.12.0-fsck.patch | 56 ++++++++++++++++++++++ 2 files changed, 93 insertions(+)
=sys-fs/f2fs-tools-1.12.0-r1 is still affected by this at least on ARM. 1.12 should remain masked, until this is resolved as it can render the whole system unavailable if rootfs is on f2fs and has the fsck flag on in fstab. Has this patch resolved the issue for the others? Which architectures? Thanks! -N
(In reply to Nikolay Kichukov from comment #14) > =sys-fs/f2fs-tools-1.12.0-r1 is still affected by this at least on ARM. 1.12 > should remain masked, until this is resolved as it can render the whole > system unavailable if rootfs is on f2fs and has the fsck flag on in fstab. > > Has this patch resolved the issue for the others? Which architectures? > > Thanks! > -N for RO-fs -f switch should be used
This seems to be the case: /etc/init.d/fsck script contains: if _forcefsck; then fsck_opts="$fsck_opts -f" check_extra="(check forced)" However it fails to boot, still. Downgrade of sys-f2/f2fs-tools fixes the issue.
My system was also rendered unusable by this. Downgrading to f2fs-tools-1.11.0-r1 fixed the problem for me, or with 1.12.0-r1, the solution is to change the last column in the fstab's root mount to 0. I'm running gentoo with kernel 4.20.17 x86_64, systemd 242-r6. This caused my system to break after an update and it remained unusable for months. Only yesterday I had the time to sit down and find a solution, I think f2fs-tools should be masked until the problem is fixed as it just renders a system unusable and it's very hard (at least it was for me) to diagnose and laborious to recover from. Even on IRC #gentoo no one was able to help.
The title of this bug has a typo in the package category (sys-f2). This should be "sys-fs". I am assuming this is why this bug is not listed when a basic search of 'sys-fs/f2fs-tools' is done. This comment seems the best way to bring this problem to the attention the right person(s). Thanks, Doug
