[opensuse-factory] BTRFS info (device dm-0): forced readonly
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339 After the Tumbleweed update yesterday I've run the btrfs balance script. Halfway thru it set the filesystem read-only. I've rebooted, it rebuild some free extent cache that it said was missing, created a new journal and I could then finish the rebalance. Today the filesystem has again hit some problem: [ 550.910590] ------------[ cut here ]------------ [ 550.910644] WARNING: CPU: 2 PID: 524 at ../fs/btrfs/extent-tree.c:6945 __btrfs_free_extent.isra.70+0x7f8/0xcc0 [btrfs] [ 550.910646] Modules linked in: fuse ip6t_REJECT nf_reject_ipv6 nf_log_ipv6 nf_conntrack_ipv6 nf_defrag_ipv6 ip6table_raw ipt_REJECT nf_reject_ipv4 xt_tcpudp xt_pkttype xt_ph ysdev br_netfilter nf_log_ipv4 nf_log_common xt_LOG xt_limit iptable_raw xt_CT iptable_filter ip6table_mangle nf_conntrack_netbios_ns nf_conntrack_broadcast nf_conntrack_ipv4 n f_defrag_ipv4 ip_tables xt_conntrack nf_conntrack ip6table_filter ip6_tables x_tables af_packet bridge stp llc iscsi_ibft iscsi_boot_sysfs msr nls_iso8859_1 nls_cp437 vfat xfs libcrc32c fat snd_hda_codec_realtek snd_hda_codec_generic intel_rapl x86_pkg_temp_thermal intel_powerclamp coretemp snd_emu10k1 snd_util_mem snd_rawmidi kvm_intel kvm snd_seq_d evice snd_ac97_codec irqbypass crct10dif_pclmul crc32_pclmul snd_hda_intel ghash_clmulni_intel snd_hda_codec [ 550.910718] aesni_intel snd_hda_core snd_hwdep pl2303 snd_pcm usbserial aes_x86_64 mei_wdt lrw iTCO_wdt i2c_i801 glue_helper iTCO_vendor_support dcdbas emu10k1_gp ablk_help er cryptd i2c_smbus pcspkr gameport ac97_bus snd_timer battery snd thermal ie31200_edac e1000e fjes tpm_tis soundcore mei_me lpc_ich tpm_tis_core ptp mei tpm pps_core edac_core shpchp mfd_core btrfs xor zlib_deflate raid6_pq crc32c_intel serio_raw sr_mod cdrom ehci_pci ehci_hcd i915 i2c_algo_bit drm_kms_helper syscopyarea xhci_pci sysfillrect sysimgb lt fb_sys_fops xhci_hcd usbcore drm usb_common button video dm_mirror dm_region_hash dm_log sg dm_multipath dm_mod scsi_dh_rdac scsi_dh_emc scsi_dh_alua efivarfs [ 550.910805] CPU: 2 PID: 524 Comm: btrfs-transacti Not tainted 4.8.6-2-default #1 [ 550.910807] Hardware name: Dell Inc. PowerEdge T20/0VD5HY, BIOS A06 01/27/2015 [ 550.910810] 0000000000000000 ffffffff833a4272 0000000000000000 0000000000000000 [ 550.910817] ffffffff8307de2e 000000948ea83000 00000000fffffffe ffff8b124d565640 [ 550.910822] ffff8b1312be2000 0000000000000000 ffff8b12893d9ee0 ffffffffc072d628 [ 550.910828] Call Trace: [ 550.910853] [<ffffffff8302eefe>] dump_trace+0x5e/0x310 [ 550.910863] [<ffffffff8302f2cb>] show_stack_log_lvl+0x11b/0x1a0 [ 550.910871] [<ffffffff83030001>] show_stack+0x21/0x40 [ 550.910880] [<ffffffff833a4272>] dump_stack+0x5c/0x7a [ 550.910889] [<ffffffff8307de2e>] __warn+0xbe/0xe0 [ 550.910927] [<ffffffffc072d628>] __btrfs_free_extent.isra.70+0x7f8/0xcc0 [btrfs] [ 550.910970] [<ffffffffc073173f>] __btrfs_run_delayed_refs+0x56f/0x12f0 [btrfs] [ 550.911006] [<ffffffffc073544a>] btrfs_run_delayed_refs+0x9a/0x2a0 [btrfs] [ 550.911047] [<ffffffffc074b3b0>] btrfs_commit_transaction+0x40/0xa80 [btrfs] [ 550.911087] [<ffffffffc07462c2>] transaction_kthread+0x1c2/0x1f0 [btrfs] [ 550.911097] [<ffffffff8309c5ed>] kthread+0xbd/0xe0 [ 550.911106] [<ffffffff836d461f>] ret_from_fork+0x1f/0x40 [ 550.915472] DWARF2 unwinder stuck at ret_from_fork+0x1f/0x40 [ 550.915475] Leftover inexact backtrace: … [ 550.916354] BTRFS error (device dm-0): unable to find ref byte nr 638048546816 parent 0 root 282 owner 30689 offset 0 [ 550.916360] ------------[ cut here ]------------ [ 550.916391] WARNING: CPU: 2 PID: 524 at ../fs/btrfs/extent-tree.c:6951 __btrfs_free_extent.isra.70+0x9ca/0xcc0 [btrfs] [ 550.916393] BTRFS: Transaction aborted (error -2) … [ 550.921107] ---[ end trace 3a5d969e50f23ffa ]--- [ 550.921111] BTRFS: error (device dm-0) in __btrfs_free_extent:6951: errno=-2 No such entry [ 550.921115] BTRFS info (device dm-0): forced readonly [ 550.921119] BTRFS: error (device dm-0) in btrfs_run_delayed_refs:2960: errno=-2 No such entry [ 550.921141] systemd-journald[586]: Failed to truncate file to its own size: Read-only file system [ 550.921219] systemd-journald[586]: Failed to truncate file to its own size: Read-only file system [ 550.921285] systemd-journald[586]: Failed to truncate file to its own size: Read-only file system [ 550.921341] systemd-journald[586]: Failed to truncate file to its own size: Read-only file system [ 550.921350] pending csums is 8192 Is there something I can do besides another reboot? Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ Waldorf MIDI Implementation & additional documentation: http://Synth.Stromeko.net/Downloads.html#WaldorfDocs -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Achim Gratz writes:
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339 … Is there something I can do besides another reboot?
Booted from a freshly created Rescue system on USB and ran btrfs check. The filesystem has missing backrefs / backpointer mismatches and/or two extent buffer leaks. Running the check with the '--mode lowmem' option produces lots of errors due to backref mismatches and also shows wrong info on the allocation statistics. The filesystem initially mounts rw, then after some time when anything hits the erroneous info it switches ro. Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ Waldorf MIDI Implementation & additional documentation: http://Synth.Stromeko.net/Downloads.html#WaldorfDocs -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
Achim Gratz writes:
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339
Is there something I can do besides another reboot?
I've played around with an image of the system-root volume and none of the repair and rescue things I've tried is actually improving things. So I need to come up with some way to re-create a fresh filesystem and populate it, preferrably with the content of what I have now. It's OK if that's without the data in the old snapshots, which is probably where the error is anyway. Can anyone tell me how to do this and/or has any pointers to some tutorials or documentation? Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ Wavetables for the Waldorf Blofeld: http://Synth.Stromeko.net/Downloads.html#BlofeldUserWavetables -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
11.11.2016 23:25, Achim Gratz пишет:
Achim Gratz writes:
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339
Is there something I can do besides another reboot?
I've played around with an image of the system-root volume and none of the repair and rescue things I've tried is actually improving things.
You expect us to second-guess what you may actually have done?
So I need to come up with some way to re-create a fresh filesystem and populate it, preferrably with the content of what I have now. It's OK if that's without the data in the old snapshots, which is probably where the error is anyway. Can anyone tell me how to do this and/or has any pointers to some tutorials or documentation?
Try btrfs restore. -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
I've played around with an image of the system-root volume and none of the repair and rescue things I've tried is actually improving things.
You expect us to second-guess what you may actually have done?
btrfs check --repair btrfs check --init-extent-tree btrfs rescue chunk-recover
Try btrfs restore.
I did (after posting my last message). I still need to figure out how to only recover the content of the 1974 snapshot that is the current root. A full restore with snapshots infinitely recurses into /.snapshots/1974/.snapshots/1974… so it will not complete. But even if I do that the other question is still how to re-initialize the volume with a sane btrfs filesystem that has all the necessary subvolumes and the same UUID. Actually I'd want to create that someplace else in an image file so I can check it before I finally try to dd it into the place of the current botched volume. Another option would probably be to do a new installation, replacing the current sysroot and picking up the /home volume. Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ SD adaptations for Waldorf Q V3.00R3 and Q+ V3.54R2: http://Synth.Stromeko.net/Downloads.html#WaldorfSDada -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
On 2016-11-12 21:16, Achim Gratz wrote:
But even if I do that the other question is still how to re-initialize the volume with a sane btrfs filesystem that has all the necessary subvolumes and the same UUID. Actually I'd want to create that someplace else in an image file so I can check it before I finally try to dd it into the place of the current botched volume.
This was asked about a week or two ago. There is no single script or
command to do it. You have to do manually all volumes yourself. Or,
install fresh again with the DVD, then erase the contents, then restore
the backup.
Subject: Re: [opensuse-factory] btrfs root - why so many subvolumes?
that was a month ago.
In particular, this post:
Date: Tue, 18 Oct 2016 10:10:11 +0200 (CEST)
From: Johannes Meixner
On 2016-11-11 21:25, Achim Gratz wrote:
Achim Gratz writes:
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339
Is there something I can do besides another reboot?
I've played around with an image of the system-root volume and none of the repair and rescue things I've tried is actually improving things. So I need to come up with some way to re-create a fresh filesystem and populate it, preferrably with the content of what I have now. It's OK if that's without the data in the old snapshots, which is probably where the error is anyway. Can anyone tell me how to do this and/or has any pointers to some tutorials or documentation?
No... the only known way is to install again. In this case, install again, then delete all, then copy from backup. -- Cheers / Saludos, Carlos E. R. (from 13.1 x86_64 "Bottle" at Telcontar)
Achim Gratz writes:
Filed as: https://bugzilla.opensuse.org/show_bug.cgi?id=1009339 […] Is there something I can do besides another reboot?
I've ended up doing a complete re-install. I had a detached /home and the installer offered to just take it as is and only re-create the system root volume without rebuilding the LVM container, just as I had intended. I now also have the new sysroot layout with the '@' subvolume (my original install was from almost exactly two years ago). Regards, Achim. -- +<[Q+ Matrix-12 WAVE#46+305 Neuron microQkb Andromeda XTk Blofeld]>+ Factory and User Sound Singles for Waldorf Blofeld: http://Synth.Stromeko.net/Downloads.html#WaldorfSounds -- To unsubscribe, e-mail: opensuse-factory+unsubscribe@opensuse.org To contact the owner, e-mail: opensuse-factory+owner@opensuse.org
participants (3)
-
Achim Gratz -
Andrei Borzenkov -
Carlos E. R.