Page 1 of 3 123 LastLast
Results 1 to 10 of 30

Thread: BtrFS has gone ReadOnly... again.

  1. #1
    Join Date
    Jun 2017
    Location
    Australia
    Posts
    440

    Exclamation BtrFS has gone ReadOnly... again.

    Hello

    For direct background... https://forums.opensuse.org/showthre...925-dup-failed

    I thought everything was hunky-dory again now [ie, ok]. Since this afternoon's Snapper Rollback, my TW Tower had been running well again. It had only been idling this evening, whilst i was off doing other stuff away from it. I'd not installed or removed any software, nor attempted any Repo-associated activity, since the successful rollback.

    It's now my 23:45, & ~30' ago i just returned to it "briefly" to do something quick, when to my shock i realised it was not working properly, again... eg:
    1. Launch YaST
      • Code:
        Configuration file "/root/.config/y2controlcenterrc" not writable.
        Please contact your system administrator.

    2. Then launch YaST2
      • Code:
        Configuration file "/root/.config/rubyrc" not writable.
        Please contact your system administrator.

    3. Attempt to refresh repos in Konsole
      • Code:
        gooeygirl@linux-Tower:~> sudo zypper refresh
        sudo: unable to open /var/lib/sudo/ts/gooeygirl: Read-only file system


    I examined KSystemLog, & found that tonight [when i was not at pc]:
    Code:
    Friday, 29 September 2017 22:30:30 AEST    kernel    BTRFS info (device sda2): forced readonly
    Oh noooo, that's also what happened just so very recently, per the linked thread above. Here's every log entry from ~an hour before it occurred tonight: https://pastebin.com/cZcaS4pN

    Sigh, i went to bed last night with a broken system, fixed it today, now going to bed again with it broken once more. Thanks universe. All comments & advice would be welcome, pls.

    PS: Now 30/9 00:15 as i Send this post.
    OS: openSUSE Tumbleweed KDE x64 Plasma 5.10.5. Browser: Vivaldi-Snapshot x64. She not he, her not him.

  2. #2
    Join Date
    Jun 2008
    Location
    Podunk
    Posts
    23,166
    Blog Entries
    15

    Default Re: BtrFS has gone ReadOnly... again.

    Hi
    Any coredumps? Maybe start running things as real root (su - ) rather than sudo?
    [CODE]
    coredumpctl list
    [/CODE
    Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890)
    openSUSE Leap 42.2 (x86_64) GNOME 3.20.2
    If you find this post helpful and are logged into the web interface,
    please show your appreciation and click on the star below... Thanks!

  3. #3
    Join Date
    Jun 2017
    Location
    Australia
    Posts
    440

    Default Re: BtrFS has gone ReadOnly... again.

    Thanks Malcolm

    I was off to bed, but thought of one more thing i wanted to check, so wandered back in & saw your post. I don't know if i did this correctly, & if so, whether anything herein is significant:
    Code:
    gooeygirl@linux-Tower:~> coredumpctl list
    Hint: You are currently not seeing messages from other users and the system.
          Users in the 'systemd-journal' group can see all messages. Pass -q to
          turn off this notice.
    No coredumps found.
    
    
    gooeygirl@linux-Tower:~> sudo coredumpctl list
    sudo: unable to open /var/lib/sudo/ts/gooeygirl: Read-only file system
    [sudo] password for root: 
    TIME                            PID   UID   GID SIG COREFILE  EXE
    Wed 2017-08-23 00:56:45 AEST  30142  1000   100  31 none      /usr/bin/pulseaudio
    Wed 2017-08-23 01:02:11 AEST   2109  1000   100  11 none      /usr/bin/plasmashell
    Wed 2017-08-23 17:28:59 AEST  14492  1000   100  31 none      /usr/bin/pulseaudio
    Wed 2017-08-23 17:57:34 AEST  23655  1000   100  31 none      /usr/bin/pulseaudio
    Wed 2017-08-23 19:57:20 AEST   2890  1000   100   6 none      /usr/bin/kactivitymanagerd
    Wed 2017-08-23 19:57:20 AEST  30954     0     0   6 none      /usr/bin/kactivitymanagerd
    Wed 2017-08-23 19:57:20 AEST   2799  1000   100  11 none      /usr/bin/ksmserver
    Wed 2017-08-23 19:57:30 AEST   3658  1000   100   6 none      /opt/teamviewer/tv_bin/TVGuiSlave.64
    Wed 2017-08-23 20:02:00 AEST   4909  1000   100   5 none      /usr/bin/clementine
    Wed 2017-08-23 20:06:57 AEST   8422     0     0  11 none      /usr/bin/ruby
    Thu 2017-08-24 00:07:03 AEST   3747  1000   100   5 none      /usr/bin/python2.7
    Thu 2017-08-24 08:27:58 AEST   6129  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Thu 2017-08-24 12:37:42 AEST  21314  1000   100  11 none      /opt/kingsoft/wps-office/office6/et
    Thu 2017-08-24 23:35:32 AEST   2786  1000   100  11 none      /usr/bin/plasmashell
    Fri 2017-08-25 12:34:53 AEST  15867  1000   100  31 none      /usr/bin/pulseaudio
    Fri 2017-08-25 15:41:41 AEST   6677  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-08-28 09:29:07 AEST   8761  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-08-28 09:29:11 AEST   2720  1000   100  11 none      /usr/bin/plasmashell
    Mon 2017-08-28 20:28:44 AEST   4583  1000   100  31 none      /usr/bin/pulseaudio
    Mon 2017-08-28 20:31:40 AEST   4792  1000   100  31 none      /usr/lib64/qt5/libexec/QtWebEngineProcess
    Mon 2017-08-28 20:31:40 AEST   4786  1000   100   6 none      /usr/bin/akregator
    Mon 2017-08-28 20:42:36 AEST   5415  1000   100  31 none      /usr/bin/pulseaudio
    Tue 2017-08-29 16:53:46 AEST  10636  1000   100   6 none      /usr/bin/firejail
    Thu 2017-08-31 13:23:14 AEST   3469  1000   100  31 none      /usr/bin/dosemu.bin
    Thu 2017-08-31 17:23:12 AEST  10202  1001   100  11 missing   /usr/bin/plasmashell
    Thu 2017-08-31 17:25:03 AEST  10268  1001   100   6 missing   /usr/bin/kactivitymanagerd
    Fri 2017-09-01 09:50:15 AEST   8762  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-02 21:42:38 AEST   4577  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-02 23:56:04 AEST   3363  1000   100   6 none      /usr/bin/cairo-dock
    Sun 2017-09-03 09:18:23 AEST   4709  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-03 11:18:48 AEST   4698  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-03 12:12:26 AEST   4695  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin                                                                  
    Sun 2017-09-03 17:10:34 AEST  12594  1000   100  31 none      /usr/bin/pulseaudio                                                                                
    Mon 2017-09-04 13:07:23 AEST    925  1000   100  31 none      /usr/bin/pulseaudio                                                                                
    Mon 2017-09-04 13:48:59 AEST   2411  1000   100  31 none      /Seagate/4. Software/Security/KeePassX/KeePassXC-2.2.0-x86_64.AppImage                             
    Mon 2017-09-04 13:50:06 AEST   2459  1000   100  31 none      /Seagate/4. Software/Security/KeePassX/KeePassXC-2.2.0-x86_64.AppImage                             
    Mon 2017-09-04 14:00:29 AEST   2887  1000   100  31 none      /home/gooeygirl/Downloads/AppImages active/KeePassXC-2.2.0-x86_64.AppImage                           
    Mon 2017-09-04 17:11:22 AEST   8378  1000   100  31 none      /home/gooeygirl/Downloads/KeePassXC-2.2.0-x86_64.AppImage                                            
    Mon 2017-09-04 18:05:32 AEST  10075  1000   100  31 none      /home/gooeygirl/Downloads/KeePassXC-2.2.0-x86_64.AppImage                                            
    Mon 2017-09-04 18:42:29 AEST  11530  1000   100   5 none      /opt/vivaldi/vivaldi-bin                                                                           
    Tue 2017-09-05 10:00:56 AEST   4710  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin                                                                  
    Wed 2017-09-06 07:38:36 AEST   4167  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Thu 2017-09-07 12:07:36 AEST  26023  1000   100  31 none      /usr/bin/pulseaudio
    Thu 2017-09-07 12:15:04 AEST  26477  1000   100  31 none      /usr/bin/pulseaudio
    Thu 2017-09-07 12:53:52 AEST  30188  1000   100  31 none      /usr/bin/pulseaudio
    Fri 2017-09-08 12:35:07 AEST   2759  1000   100  11 none      /usr/bin/plasmashell
    Fri 2017-09-08 13:21:05 AEST   6265  1001   100   6 missing   /usr/bin/kactivitymanagerd
    Fri 2017-09-08 14:32:21 AEST   2735  1000   100  11 none      /usr/bin/ksmserver
    Fri 2017-09-08 14:32:21 AEST   2825  1000   100   6 none      /usr/bin/kactivitymanagerd
    Sat 2017-09-09 09:01:19 AEST  11393  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-09-11 23:04:05 AEST   3865  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-09-11 23:54:26 AEST   3868  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Tue 2017-09-12 11:22:47 AEST   3779  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Wed 2017-09-13 08:48:37 AEST   4346  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Wed 2017-09-13 09:01:59 AEST   4347  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Thu 2017-09-14 12:38:00 AEST  23307  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Thu 2017-09-14 18:59:59 AEST  14936  1000   100  31 none      /usr/bin/pulseaudio
    Fri 2017-09-15 14:19:07 AEST   4165  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-15 14:19:13 AEST   3883  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-16 08:09:09 AEST   4493  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-16 19:43:09 AEST   4018  1001   100  11 missing   /usr/bin/plasmashell
    Sat 2017-09-16 20:49:46 AEST  20975  1001   100   5 missing   /usr/bin/python2.7
    Sat 2017-09-16 20:49:46 AEST   4075  1001   100   6 missing   /usr/bin/kactivitymanagerd
    Sun 2017-09-17 06:53:17 AEST  24691  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-17 07:01:34 AEST  24696  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-17 07:18:05 AEST  24654  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-17 14:25:22 AEST  31118  1000   100  31 none      /usr/bin/pulseaudio
    Mon 2017-09-18 09:04:40 AEST  24651  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-09-18 16:32:09 AEST   2745  1000   100  11 none      /usr/bin/plasmashell
    Tue 2017-09-19 07:20:05 AEST   5871  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Tue 2017-09-19 18:53:21 AEST  16877  1000   100  31 none      /usr/bin/pulseaudio
    Wed 2017-09-20 08:39:39 AEST  28532  1000   100   5 none      /opt/vivaldi-snapshot/vivaldi-bin
    Wed 2017-09-20 14:33:00 AEST  15211  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Wed 2017-09-20 16:42:50 AEST   3125  1000   100   5 none      /usr/bin/clementine
    Fri 2017-09-22 09:36:10 AEST    980  1000   100  31 none      /usr/lib64/thunderbird/thunderbird-bin
    Fri 2017-09-22 09:43:43 AEST  31728  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 13:42:28 AEST  31482  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 15:41:29 AEST  14633  1000   100   6 none      /usr/bin/firejail
    Fri 2017-09-22 19:01:15 AEST  21000  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:01:16 AEST  21011  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:01:33 AEST  21026  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:01:34 AEST  21035  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:02:59 AEST  21934  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:02:59 AEST  21945  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:03:23 AEST  21981  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:03:23 AEST  21987  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:06:09 AEST  22173  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:06:09 AEST  22179  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:06:32 AEST  22205  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:32:20 AEST  25887  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:32:21 AEST  25895  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:32:38 AEST  25932  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Fri 2017-09-22 19:32:39 AEST  25941  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-23 07:53:20 AEST   6215  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sat 2017-09-23 11:14:38 AEST  16373  1000   100  31 none      /usr/bin/pulseaudio
    Sat 2017-09-23 13:07:01 AEST   8205  1000   100  11 none      /usr/bin/perl
    Sat 2017-09-23 13:46:57 AEST  23382  1000   100   6 none      /usr/bin/firejail
    Sun 2017-09-24 08:42:31 AEST   5755  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Sun 2017-09-24 15:06:43 AEST  13254     0     0   6 none      /usr/bin/ruby
    Sun 2017-09-24 16:48:02 AEST  17226     4     7  11 missing   /usr/local/Brother/cupswrapper/brcupsconfig3
    Sun 2017-09-24 17:23:54 AEST  20427     0     0   6 none      /usr/bin/ruby
    Sun 2017-09-24 17:28:58 AEST  23762  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 17:30:58 AEST  23846  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 20:01:49 AEST   4045  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 20:02:55 AEST   4074  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 20:07:53 AEST   4093  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 20:08:05 AEST   4138  1000   100  31 none      /usr/bin/pulseaudio
    Sun 2017-09-24 20:22:58 AEST   5938  1000   100  31 none      /usr/bin/pulseaudio
    Mon 2017-09-25 10:52:06 AEST   4620  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-09-25 12:41:47 AEST   4629  1000   100  11 none      /opt/vivaldi-snapshot/vivaldi-bin
    Mon 2017-09-25 19:34:07 AEST  18031  1000   100  31 none      /usr/bin/pulseaudio
    Mon 2017-09-25 19:35:23 AEST  18083  1000   100  31 none      /usr/bin/pulseaudio
    Tue 2017-09-26 12:06:19 AEST   3797  1000   100  31 none      /usr/bin/pulseaudio
    Tue 2017-09-26 13:04:45 AEST  11002  1000   100  31 none      /usr/bin/pulseaudio
    Tue 2017-09-26 13:44:10 AEST   3532  1001   100  31 missing   /usr/bin/pulseaudio
    Tue 2017-09-26 13:44:22 AEST   3578  1001   100  31 missing   /usr/bin/pulseaudio
    Tue 2017-09-26 19:47:05 AEST   3760  1000   100  11 none      /usr/bin/plasmashell
    Tue 2017-09-26 19:54:30 AEST  18830  1000   100  11 none      /usr/bin/plasmashell
    Thu 2017-09-28 16:20:19 AEST  15649  1000   100  31 none      /usr/bin/pulseaudio
    Thu 2017-09-28 16:22:11 AEST  15794  1000   100  31 none      /usr/bin/pulseaudio
    Thu 2017-09-28 17:05:16 AEST  16740  1000   100  31 none      /usr/bin/pulseaudio
    Fri 2017-09-29 12:09:09 AEST   3965  1000   100   5 none      /usr/bin/python2.7
    lines 75-123/123 (END)

    PS: I thought it was regarded as Sehr Verboten to run stuff as real root...?
    OS: openSUSE Tumbleweed KDE x64 Plasma 5.10.5. Browser: Vivaldi-Snapshot x64. She not he, her not him.

  4. #4
    Join Date
    Jun 2017
    Location
    Australia
    Posts
    440

    Default Re: BtrFS has gone ReadOnly... again.

    This is that other thing i wanted to check. Way back when i first installed TW, with the Ruby Installer i set the No Access Time flag on my partitions [it was a legacy habit of mine from pre-oS, supposedly to enhance the SSD life]. If i recall [& i'm just about falling down tired, so this might be wrong], way back you [or maybe Henk?] queried me on why i'd done that, & mentioned that with modern SSDs it was redundant. Back then i decided that "redundant" was not the same as "bad", so i left it as-is, then forgot about it. Some time later [but still a fair while back now], i think i read somewhere that using noatime with BtrFS might actually lead to BtrFS becoming less reliable, & possibly resulting in data corruption/loss [again, i'm tired, that might be hogwash].

    So the question is, wrt this now twice-occurred BtrFS resetting to ReadOnly... might noatime be a factor, & should i now remove all of those entries from my fstab?
    Code:
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c / btrfs noatime 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /boot/grub2/i386-pc btrfs noatime,subvol=@/boot/grub2/i386-pc 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /boot/grub2/x86_64-efi btrfs noatime,subvol=@/boot/grub2/x86_64-efi 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /opt btrfs noatime,subvol=@/opt 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /srv btrfs noatime,subvol=@/srv 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /usr/local btrfs noatime,subvol=@/usr/local 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/cache btrfs noatime,subvol=@/var/cache 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/crash btrfs noatime,subvol=@/var/crash 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/libvirt/images btrfs noatime,subvol=@/var/lib/libvirt/images 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/machines btrfs noatime,subvol=@/var/lib/machines 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mailman btrfs noatime,subvol=@/var/lib/mailman 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mariadb btrfs noatime,subvol=@/var/lib/mariadb 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mysql btrfs noatime,subvol=@/var/lib/mysql 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/named btrfs noatime,subvol=@/var/lib/named 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/pgsql btrfs noatime,subvol=@/var/lib/pgsql 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/log btrfs noatime,subvol=@/var/log 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/opt btrfs noatime,subvol=@/var/opt 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/spool btrfs noatime,subvol=@/var/spool 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/tmp btrfs noatime,subvol=@/var/tmp 0 0
    /dev/mapper/cr_ata-ST2000DM001-1ER164_Z8E001EQ-part1 swap swap defaults 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /.snapshots btrfs noatime,subvol=@/.snapshots 0 0
    UUID=23927deb-03f4-4b7b-9599-9e44c9f86919 /Seagate             ext4       noatime,acl           1 2
    UUID=f4ae6a8c-a541-4bbd-b086-6358872a3962 /SeagateSpare        xfs        noatime               1 2
    UUID=DFF3-1AD7       /boot/efi            vfat       umask=0002,utf8=true  0 0
    /dev/mapper/cr_ata-Samsung_SSD_850_EVO_250GB_S21MNSAG105383J-part3 /home                xfs        noatime,nofail        0 2
    tmpfs                /tmp                 tmpfs      size=2G               0 0
    Finally, is it safe to reboot now in this ReadOnly status? Will that reset with the reboot, or is it now "stuck" like that?

    I'll read your anticipated [thanks] response after i wake up again.
    OS: openSUSE Tumbleweed KDE x64 Plasma 5.10.5. Browser: Vivaldi-Snapshot x64. She not he, her not him.

  5. #5
    Join Date
    Jun 2008
    Location
    Podunk
    Posts
    23,166
    Blog Entries
    15

    Default Re: BtrFS has gone ReadOnly... again.

    Hi
    Quite a few coredumps....

    You need to inspect with coredumpctl gdb <PID>

    Could be noatime related, always used the defaults for awhile now, also running the bfq scheduler on Tumbleweed (with a SSD), system is rock solid.
    Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890)
    openSUSE Leap 42.2 (x86_64) GNOME 3.20.2
    If you find this post helpful and are logged into the web interface,
    please show your appreciation and click on the star below... Thanks!

  6. #6
    Join Date
    Jun 2008
    Location
    Podunk
    Posts
    23,166
    Blog Entries
    15

    Default Re: BtrFS has gone ReadOnly... again.

    Quote Originally Posted by GooeyGirl View Post
    This is that other thing i wanted to check. Way back when i first installed TW, with the Ruby Installer i set the No Access Time flag on my partitions [it was a legacy habit of mine from pre-oS, supposedly to enhance the SSD life]. If i recall [& i'm just about falling down tired, so this might be wrong], way back you [or maybe Henk?] queried me on why i'd done that, & mentioned that with modern SSDs it was redundant. Back then i decided that "redundant" was not the same as "bad", so i left it as-is, then forgot about it. Some time later [but still a fair while back now], i think i read somewhere that using noatime with BtrFS might actually lead to BtrFS becoming less reliable, & possibly resulting in data corruption/loss [again, i'm tired, that might be hogwash].

    So the question is, wrt this now twice-occurred BtrFS resetting to ReadOnly... might noatime be a factor, & should i now remove all of those entries from my fstab?
    Code:
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c / btrfs noatime 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /boot/grub2/i386-pc btrfs noatime,subvol=@/boot/grub2/i386-pc 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /boot/grub2/x86_64-efi btrfs noatime,subvol=@/boot/grub2/x86_64-efi 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /opt btrfs noatime,subvol=@/opt 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /srv btrfs noatime,subvol=@/srv 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /usr/local btrfs noatime,subvol=@/usr/local 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/cache btrfs noatime,subvol=@/var/cache 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/crash btrfs noatime,subvol=@/var/crash 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/libvirt/images btrfs noatime,subvol=@/var/lib/libvirt/images 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/machines btrfs noatime,subvol=@/var/lib/machines 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mailman btrfs noatime,subvol=@/var/lib/mailman 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mariadb btrfs noatime,subvol=@/var/lib/mariadb 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/mysql btrfs noatime,subvol=@/var/lib/mysql 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/named btrfs noatime,subvol=@/var/lib/named 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/lib/pgsql btrfs noatime,subvol=@/var/lib/pgsql 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/log btrfs noatime,subvol=@/var/log 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/opt btrfs noatime,subvol=@/var/opt 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/spool btrfs noatime,subvol=@/var/spool 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /var/tmp btrfs noatime,subvol=@/var/tmp 0 0
    /dev/mapper/cr_ata-ST2000DM001-1ER164_Z8E001EQ-part1 swap swap defaults 0 0
    UUID=f3e11c85-7f1a-4e9e-8585-5b6a61e4ea8c /.snapshots btrfs noatime,subvol=@/.snapshots 0 0
    UUID=23927deb-03f4-4b7b-9599-9e44c9f86919 /Seagate             ext4       noatime,acl           1 2
    UUID=f4ae6a8c-a541-4bbd-b086-6358872a3962 /SeagateSpare        xfs        noatime               1 2
    UUID=DFF3-1AD7       /boot/efi            vfat       umask=0002,utf8=true  0 0
    /dev/mapper/cr_ata-Samsung_SSD_850_EVO_250GB_S21MNSAG105383J-part3 /home                xfs        noatime,nofail        0 2
    tmpfs                /tmp                 tmpfs      size=2G               0 0
    Finally, is it safe to reboot now in this ReadOnly status? Will that reset with the reboot, or is it now "stuck" like that?

    I'll read your anticipated [thanks] response after i wake up again.
    Hi
    So if you run the command mount what are all the options used? I guess it would reboot, but maybe back to RO... I would be tempted to remove your noatime from btrfs mounts.
    Cheers Malcolm °¿° SUSE Knowledge Partner (Linux Counter #276890)
    openSUSE Leap 42.2 (x86_64) GNOME 3.20.2
    If you find this post helpful and are logged into the web interface,
    please show your appreciation and click on the star below... Thanks!

  7. #7
    Join Date
    Sep 2013
    Location
    Norfolk, UK
    Posts
    602

    Default Re: BtrFS has gone ReadOnly... again.

    Quote Originally Posted by malcolmlewis View Post
    I would be tempted to remove your noatime from btrfs mounts.
    Things may well have changed since theses two articles, which are approx 5 years old, but they advocate the use of noatime as the default on btrfs.

    https://lwn.net/Articles/499293/
    https://lwn.net/Articles/499294/

    And if one scrolls down to almost the last paragraph on this much more recent (2017) wiki on the kernel.org site it also suggests using noatime.

    https://btrfs.wiki.kernel.org/index.php/Mount_options

    So perhaps the OP's problem isn't noatime related and lies elsewhere.
    Regards, Paul

    Tumbleweed (Snapshot: 20171017) KDE Plasma 5
    Leap 42.3 KDE Plasma 5

  8. #8
    Join Date
    Sep 2012
    Posts
    4,046

    Default Re: BtrFS has gone ReadOnly... again.

    Let's face it - your filesystem is corrupted. It may work for some time until it hits bad spot again when it screams. You can try btrfs repair, but honestly, if your goal is to get working system back - recreate filesystem and restore from backup.

  9. #9

    Default Re: BtrFS has gone ReadOnly... again.

    Your system is corrupted and now unreliable. You need to start over, it's too far gone.

    I would suggest reinstalling a temporary Linux or Windows, and doing some stress testing first for a couple days with Prime95, IBT, compile loops, AIDA64, MemTest, etc. See if you can catch any cpu/mem hardware glitches.

    Also, try to verify that your ssd / hdd is not on the way out.

  10. #10
    Join Date
    Jun 2017
    Location
    Australia
    Posts
    440

    Default Re: BtrFS has gone ReadOnly... again.

    Thanks for all replies, rather depressing though the tone is.

    Qualifier before i ask my undoubtedly naive question below... I am not disputing anyone's suggestion that my system is apparently beyond repair [though i still can't grasp how such calamity should have arisen], but it seems i still haven't understood something basic to the whole Tumbleweed zypper dup concept.

    Let's face it - your filesystem is corrupted. It may work for some time until it hits bad spot again when it screams
    Your system is corrupted and now unreliable. You need to start over, it's too far gone
    I had thought that the whole point of TW being a rolling release, & the approved way to "update" it being actually to "upgrade" it via zypper dup, was that [i'm sure i've read this many times across the forum] each time a virtually brand new OS is installed, replacing the previous snapshot. Obviously that is patently untrue for those occasional quite small dups that only contain a few packages, but conversely my experience frequently has been that many hundreds & indeed more than one thousand packages are involved, comprising often one or a few GB.

    Thus, even if my Tower incomprehensively somehow managed to irreparably screw up its current snapshot, wouldn't / shouldn't / doesn't the next "big" dup automagically take care of that problem by means of overwriting all the previous system files, thus making my TW "good" again?

    Furthermore, given i use BtrFS, & the other day [per my linked thread] i did an apparently successful Snapper Rollback to a snapshot prior to "all hell breaking loose", shouldn't that have ipso facto given me a "good" system once again?

    As you can see, i'm clearly badly misunderstanding the situation.


    Now, moving onto the specifics of what i need to do, despite me still not understanding the why,
    recreate filesystem and restore from backup
    Sorry, but what does this actually mean?
    1. How do i recreate my filesystem; is that just some shorthand way of saying i need to do a fresh installation of TW from media?
    2. What do i restore from backup? If this refers to all my docs & data, then of course that's fine & easy. But is it referring to "my" stuff, or to "system" stuff?


    Regarding:
    a temporary Linux ..., and doing some stress testing first for a couple days with Prime95, IBT, compile loops, AIDA64, MemTest, etc. See if you can catch any cpu/mem hardware glitches
    ,,,thank you. Later i shall DDG those things to try to decide what to do.

    With respect to my earlier question about noatime, in addition to Paul's interesting advice, i'm thinking for now that my query on that might be a red-herring [at least in terms of my current problem], because i took the same decision on my Lappy [this thread is for my Tower], on which i installed TW a few weeks before my Tower a few months ago, yet Lappy continues to be fine [touch wood**].


    **As i write this [on Tower], Lappy is also on my desk near me, & is doing its weekly zypper dup... from 20170913 to 20170928. It has just frozen at this stage [it will not respond to the "r" option, it will not launch Dolphin or TeamViewer (hence this photo rather than copied text in a codebox) -- note that Lappy was working just fine prior to initiating this dup]:



    I do not know if it was the same error, but seeing this now on Lappy has reminded me that, during a dup within the timeframe of my recent Tower dramas, it also threw a similar-looking error [ie, red font]. I can't recall what option i chose then, but i'm now wondering if:
    1. That might have lead to my Tower's TW pain?
    2. Is my Lappy TW now about to be broken as well?



    If the upshot of all this is that yes i have to do a clean install of an OS into my root partition, then... if i stay with oS... I need to decide [again] on TW or Leap, AND on [for root] BtrFS or Ext4. I have no idea if BtrFS is causally implicated in my troubles [I know that some people seem to energetically dislike BtrFS for perceived reliability reasons]. If i decide to reinstall TW, & retain BtrFS, & later suffer the same problems, i would then feel foolish that i used BtrFS again. Conversely if i reinstall TW then use ext4 for root, i am throwing away the substantial safety-net that is Snapper Rollbacks. Using a cutting edge rolling release like TW without Snapper... would that be brave or stupid of me? Or, if i decide that despite my great like for TW i can't tolerate this ongoing stress & time-stealing it's causing me, & so install Leap instead, then i am accepting an old Plasma version, & old pgms [& old kernel, though maybe that's least important to me?]. I really strongly like Plasma 5.10.x over 5.8.x, so that would be a very sad outcome for me.

    This is all really hard.
    OS: openSUSE Tumbleweed KDE x64 Plasma 5.10.5. Browser: Vivaldi-Snapshot x64. She not he, her not him.

Page 1 of 3 123 LastLast

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •