meta data for this page
  •  

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision		Previous revision
linux:fs:btrfs_as_root2 [2017/01/20 13:40] – [Replace bad disc] niziaklinux:fs:btrfs_as_root2 [2025/03/24 08:32] (current) – ↷ Links adapted because of a move operation niziak
Line 52: Line 52:
  
 **NOTE:** Create subvolumes for things which shouldn't be included in snapshot of parent volume. Like some /var/../cache, etc **NOTE:** Create subvolumes for things which shouldn't be included in snapshot of parent volume. Like some /var/../cache, etc
 +
 +[[https://wiki.archlinux.org/index.php/Snapper#Suggested_filesystem_layout]]
  
 <code bash> <code bash>
Line 119: Line 121:
  
 Create subvolumes: Create subvolumes:
-<code>+<code bash>
 cd /mnt cd /mnt
 btrfs subvolume create @pool btrfs subvolume create @pool
Line 130: Line 132:
 </file> </file>
  
 +[[https://btrfs.wiki.kernel.org/index.php/SysadminGuide#Btrfs_on_top_of_dmcrypt]]
 ===== Add 2nd disc ===== ===== Add 2nd disc =====
-Copy partition layouts from sdb to sda drive: +See [[.btrfs:multidisk]]
-<code bash> +
-sudo sfdisk -d /dev/sdb > parts +
-sudo sfdisk  /dev/sda < parts +
-</code> +
- +
-Recreate LVM layout. +
- +
-Add 2nd disc to btrfs: +
-<code bash> +
-btrfs device add /dev/disc1/btrfs1 /  +
-btrfs device usage / +
-btrfs balance start -v -dconvert=raid1 -mconvert=raid1 / +
-btrfs device usage / +
-</code> +
-Make sure that data on both disc are the same. Especially system data are on boths disc in RAID1 mode. +
- +
- +
-Let system boot without one disc: +
-<file | /etc/default/grub> +
-GRUB_CMDLINE_LINUX_DEFAULT="rootflags=degraded" +
-</file> +
-<code bash>update-grub</code> +
- +
-Install grub +
-<code bash> +
-sudo grub-install /dev/sdb +
-sudo grub-install /dev/sdc +
-reboot +
-</code> +
- +
-==== Issues ==== +
- +
-=== Grub === +
-**PROBLEM** +
-**''grub-pc'' 2.02~beta2-22+deb8u1 0** incorrectly handle multiple roots. Internal variable ${GRUB_DEVICE} which is used to generate ''root='' parameter include newline character, so grub menuconfig entry is broken: +
-<code> +
-       linux<->/@/@root/boot/vmlinuz-4.8.0-0.bpo.2-amd64 root=/dev/mapper/disc2-btrfs2 +
-/dev/mapper/disc1-btrfs1 ro rootflags=subvol=@/@root  rootflags=degraded +
-</code> +
-More here: [[https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1238347]] and [[https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/1582811]] +
-**SOLUTION** +
-Fixed in Debian Stretch (testing) version **grub-pc_2.02~beta3-3** +
-Now UUID is used multiple roots. UUID of btrfs filesystem placed on 2 disc is the same: +
-<code bash> +
-/dev/mapper/disc2-btrfs2: UUID="52944cdd-f8d0-4798-bd1c-80539c45253d" UUID_SUB="54703ddb-5789-4cac-bfd0-691acadfa33c" TYPE="btrfs" +
-/dev/mapper/disc1-btrfs1: UUID="52944cdd-f8d0-4798-bd1c-80539c45253d" UUID_SUB="9bc49b38-c014-40e3-876e-08f6873293b8" TYPE="btrfs" +
-</code> +
-So, kernel cmdline parameter **''root=UUID=...''** is now correct. +
- +
-=== no symbol table === +
-**PROBLEM** +
-Grub says: +
-<code>grub error: no symbol table</code> +
-**SOLUTION** +
-Grub was not reinstalled after update +
-<code bash> +
-grub-install /dev/sda +
-grub-install /dev/sdb +
-update-grub +
-</code> +
- +
-=== ALERT! /dev/disk/by-uuid/xxxxxxxxx does not exist. Dropping to a shell === +
- +
-**PROBLEM** +
-Grub loads kernel and initramfs correctly, but: +
-''Begin: Waiting for root file system ...'' and Grub still boot into initramfs. +
-After some while, initramfs shell is available. +
-<code> +
-ALERT!  +
-Dropping to shell! +
-</code> +
- +
-Unfortunatelly initramfs cannot activate LVM volumes when kernel cmdline ''root=UUID=...'' is used. +
-<code bash> +
-lvm lvscan +
-</code> +
-To boot manually: +
-<code bash> +
-lvm vgchange -ay +
-mount /dev/disc2/btrfs2 /root -o device=/dev/disc1/btrfs1 +
-exit +
-</code> +
-System should start - but once. +
- +
-Problem is located in ''/usr/share/initramfs-tools/scripts/local-top/lvm2''. +
-Script check if specified root device needs to be activated by LVM. +
-When UUID is used it is executing code: +
-<code bash> +
-/dev/*/*) +
-        # Could be /dev/VG/LV; use lvs to check +
-        if lvm lvs -- "$dev" >/dev/null 2>&1; then +
-                lvchange_activate "$dev" +
-        fi +
-        ;; +
-</code> +
-In result, command +
-<code bash>lvm lvs -- disc/by-uuid/52944cdd-f8d0-4798-bd1c-80539c45253d<code> +
-is executed, resulting following output: +
-<code>"disc/by-uuid/52944cdd-f8d0-4798-bd1c-80539c45253d": Invalid path for Logical Volume.</code> +
- +
-Debian bugs about similar problem: +
-  * [[https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=612402]] +
-  * [[https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=741342]] +
- +
- +
-**solution** +
- +
-Install lvm2 package version 2.02.168-1 (Debian Stretch). Previously was 2.02.111-2.2+deb8u1 (Debian Jessie). +
-<code bash> +
-update-initramfs -k all -u +
-btrfs scrub / +
-</code> +
- +
-or +
-force LVM activation: [[http://askubuntu.com/a/834626|Can't find LVM root dropped back to initramfs]] +
- +
-Also ''GRUB_DISABLE_OS_PROBER=true'' can be added. +
- +
- +
-<file | /etc/mkinitcpio.conf> +
-BINARIES="/bin/btrfs" +
-</file> +
-<code bash>update-init -u -k all</code> +
- +
-===== Disc fail (removed) ===== +
- +
-=== error: disk not found === +
-<code> +
-error: disk 'lvmid/8vc3Xl-....' not found. +
-Entering rescue mode... +
-grub rescue> +
-</code> +
- +
-GRUB 2 is unable to find the grub folder or its contents are missing/corrupted. The grub folder contains the GRUB 2 menu, modules and stored environmental data.  +
- +
-grub.cfg sets prefix variable to missing LVM (lvimd/<VG UUID/<LV UUID>), and then call to load module which fails: +
-<code> +
-prefix='(lvmid/8vc3Xl-a40o-ILx2-AoRf-9Z37-4h51-jq0OR5/Hr8CPf-dVqs-uIm1-pkfH-fJdI-14io-rwtx3l)/@/@root/boot/grub' +
-insmod gettext +
-</code> +
- +
-Rescue shell is very limited. Try to start ''normal'' shell. +
-<code> +
-set prefix='(lvm/disc1-btrfs1)/@/@root/boot/grub' +
-insmod normal +
-normal +
-</code> +
- +
- +
- +
- +
- +
- +
- +
- +
- +
-===== Replace bad disc ===== +
- +
-<code bash> +
-btrfs device add /dev/disc3/btrfs3 / +
-btrfs device delete missing / +
-btrfs balance start -v -dconvert=raid1 -mconvert=raid1 / +
-</code> +
- +
-<code bash> +
-btrfs filesystem show +
-btrfs replace start 6 /dev/disc3/btrfs3 /  # change 6 to your setup +
-</code> +
- +
-==== Fine tuning ==== +
-  * Add your ssh public key to ''/etc/initramfs-tools/root/.ssh/authorized_keys'' +
-  * Change MAC address of eth0 (for initramfs remote access) +
-<file | /etc/udev/rules.d/75-mac-spoof.rule> +
-ACTION=="add", SUBSYSTEM=="net", ATTR{address}=="08:00:27:f9:3d:3e", RUN+="/sbin/ip link set dev %k address 08:00:27:f9:12:34" +
-</file> +
-Another option is to add ''local-top'' script like described here for IPv6 [[https://www.danrl.com/2015/10/21/debian-jessi-ssh-fde-unlock.html]]+
  
-  *  
  
-==== TODO ==== 
-  * backup GPT header 
-  * backup LVM header 
-  * backup LUKS header