BTRFS RAID1 backup issues

[MVKozlov]

Relax-and-Recover 2.6 / 2020-06-17
os

NAME="Ubuntu"
VERSION="20.04.2 LTS (Focal Fossa)"
ID=ubuntu
ID_LIKE=debian
PRETTY_NAME="Ubuntu 20.04.2 LTS"
VERSION_ID="20.04"
HOME_URL="https://www.ubuntu.com/"
SUPPORT_URL="https://help.ubuntu.com/"
BUG_REPORT_URL="https://bugs.launchpad.net/ubuntu/"
PRIVACY_POLICY_URL="https://www.ubuntu.com/legal/terms-and-policies/privacy-policy"
VERSION_CODENAME=focal
UBUNTU_CODENAME=focal

site.coinf

OUTPUT=ISO
OUTPUT_URL=cifs://1.2.3.4/Backup$
OUTPUT_OPTIONS="cred=/etc/rear/cifs_credentials"

REQUIRED_PROGS+=( lsblk efibootmgr unshare tune2fs resize2fs nslookup grub-install update-grub ldconfig.real )

BACKUP=NETFS
BACKUP_URL=cifs://1.2.3.4/Backup$

BACKUP_PROG_EXCLUDE+=( '/var/lib/pgpro/main/data/base' '/var/lib/pgpro/main_base' )
BACKUP_PROG_INCLUDE+=( '/var/lib/pgpro/main_base/1' )

BACKUP_OPTIONS="cred=/etc/rear/cifs_credentials"

AUTOSHRINK_DISK_SIZE_LIMIT_PERCENTAGE=10
AUTORESIZE_PARTITIONS=""

UEFI

lsblk

NAME             KNAME          PKNAME       TRAN   TYPE  FSTYPE              SIZE MOUNTPOINT
/dev/sda         /dev/sda                    sata   disk                    894.3G
`-/dev/sda1      /dev/sda1      /dev/sda            part  btrfs             894.3G /btrfs/ssd
/dev/sdb         /dev/sdb                    sata   disk                    894.3G
`-/dev/sdb1      /dev/sdb1      /dev/sdb            part  btrfs             894.3G
/dev/sdc         /dev/sdc                    sata   disk                      5.5T
|-/dev/sdc1      /dev/sdc1      /dev/sdc            part  vfat                512M
|-/dev/sdc2      /dev/sdc2      /dev/sdc            part  linux_raid_member   197G
| `-/dev/md0     /dev/md0       /dev/sdc2           raid1                   196.9G
|   |-/dev/md0p1 /dev/md0p1     /dev/md0            part  ext4                  1G /boot
|   |-/dev/md0p2 /dev/md0p2     /dev/md0            part  ext4                 64G /
|   `-/dev/md0p3 /dev/md0p3     /dev/md0            part  swap              131.9G [SWAP]
`-/dev/sdc3      /dev/sdc3      /dev/sdc            part  btrfs                 5T /btrfs/hdd
/dev/sdd         /dev/sdd                    sata   disk                      5.5T
|-/dev/sdd1      /dev/sdd1      /dev/sdd            part  vfat                512M /boot/efi
|-/dev/sdd2      /dev/sdd2      /dev/sdd            part  linux_raid_member   197G
| `-/dev/md0     /dev/md0       /dev/sdd2           raid1                   196.9G
|   |-/dev/md0p1 /dev/md0p1     /dev/md0            part  ext4                  1G /boot
|   |-/dev/md0p2 /dev/md0p2     /dev/md0            part  ext4                 64G /
|   `-/dev/md0p3 /dev/md0p3     /dev/md0            part  swap              131.9G [SWAP]
`-/dev/sdd3      /dev/sdd3      /dev/sdd            part  btrfs                 5T
/dev/sr0         /dev/sr0                    sata   rom                      1024M
/dev/nvme0n1     /dev/nvme0n1                nvme   disk                    931.5G
`-/dev/nvme0n1p1 /dev/nvme0n1p1 /dev/nvme0n1 nvme   part  btrfs             931.5G /btrfs/nvme
/dev/nvme1n1     /dev/nvme1n1                nvme   disk                    931.5G
`-/dev/nvme1n1p1 /dev/nvme1n1p1 /dev/nvme1n1 nvme   part  btrfs             931.5G

/etc/fstab:

LABEL=raid1nvme	/mnt/btrfs/nvme	btrfs	rw,noatime,defaults	0 0
LABEL=raid1ssd	/mnt/btrfs/ssd   	btrfs	rw,noatime,defaults	0 0
LABEL=raid1hdd	/mnt/btrfs/hdd	btrfs	rw,noatime,defaults	0 0

LABEL=raid1ssd	/var/lib/pgpro/main		btrfs	subvol=postgres,rw,noatime,defaults	0 0
LABEL=raid1nvme	/var/lib/pgpro/pg_wal		btrfs	subvol=pg_wal,rw,noatime,defaults	0 0
LABEL=raid1nvme	/var/lib/pgpro/main_base	btrfs	subvol=main_base,rw,noatime,defaults	0 0
LABEL=raid1hdd	/opt				btrfs	subvol=opt,rw,noatime,defaults		0 0
LABEL=raid1hdd	/media/backup			btrfs	subvol=backup,rw,noatime,defaults	0 0
LABEL=raid1hdd	/home/usr1     			btrfs	subvol=usr1,rw,noatime,defaults	0 0

when default btrfs volumes mounted as showed in fstab (under /mnt/btrfs/xxx)
then ALL brtfs volumes skipped according to rear.log
example:

Disk /dev/nvme0n1 is not used by any mounted filesystem. Excluding.
Disk /dev/nvme1n1 is not used by any mounted filesystem. Excluding.
Disk /dev/sda is not used by any mounted filesystem. Excluding.
Disk /dev/sdb is not used by any mounted filesystem. Excluding.

when default volumes mounted as /btrfs/xxx
then only mirror disk volumes skipped in logs (sdb and nvme1n1)

but anyway nothing from any btrfs fs path go into backup

when I try to restore backup to test VM (add 6 drives sda...sdf)
only sdc and sdd volumes asks for remapping. so any btrfs data not restored at all
I try to manually add some volumes to remap and layout and restore fails on disk layout creation
(may be nvme - sdX remap is another problem)

I understand that btrfs support limited but is there any workarounds for full backup/restoration ?

[jsmeix]

A quick reply as far as I know off the top of my head:

Real hardware RAID is supported by ReaR
because underlying lower level things below the RAID do not appear.

Traditional RAID1 via MD devices aka Linux Software RAID is supported by ReaR.

Other forms of RAID like so called "Firmware RAID"
cf. #1540
or using one single btrfs filesystem on multiple devices
cf. https://btrfs.wiki.kernel.org/index.php/Using_Btrfs_with_Multiple_Devices
are currently not supported by ReaR.

[jsmeix]

Regarding btrfs subvolumes and the backup of the files see
#2604 (comment)

In general including/excluding components like disks/partitions/filesystems
is different from what gets included/excluded regarding the backup.
There are some relationships but in general both are different things,
cf. #2586 (comment)

[MVKozlov]

disklayout.conf contain uncommented btrfs drives, partitions, fs, subvolumes and mounted subvolumes and my config contain
BACKUP_PROG_INCLUDE+=( '/var/lib/pgpro/main_base/1' )

if at least the data got into the backup, this would not be a big problem, but they are silently skipped.

I feel it as a bug

[jsmeix]

Regarding the backup with BACKUP=NETFS we have
BACKUP_ONLY_INCLUDE and BACKUP_ONLY_EXCLUDE
that should help to specify explicitly what gets included in the backup.
See its documentation in the default.conf file
https://github.com/rear/rear/blob/master/usr/share/rear/conf/default.conf

The script that by default (i.e. unless BACKUP_ONLY_EXCLUDE is 'true')
implicitly also adds excluded mountpoints to the backup exclude list is
backup/NETFS/default/400_create_include_exclude_files.sh
https://github.com/rear/rear/blob/master/usr/share/rear/backup/NETFS/default/400_create_include_exclude_files.sh
and indeed it does that silently.

But what is plobably worse is that the actual backup script
backup/NETFS/default/500_make_backup.sh
https://github.com/rear/rear/blob/master/usr/share/rear/backup/NETFS/default/500_make_backup.sh
only writes the "Backup include list" and the "Backup exclude list"
to the log file so that the user won't see anything on his terminal
also not in verbose -v mode (or in debug modes -d or -D).

[MVKozlov]

thanks, a careful reading of the entire thread # 2604 helped to add all the data into backup
but it seems simultaneous
BACKUP_PROG_EXCLUDE+=( '/somedirectory' )
BACKUP_PROG_INCLUDE+=( '/somedirectory/subdirectory' )
cannot include /somedirectory/subdirectory' in backup. it not clear from the doc

for brtfs recreation I try to use
PRE_RECOVERY_SCRIPT=
POST_RECOVERY_SCRIPT=

[github-actions]

Stale issue message