1. The nbd module needs to reloaded to ensure nbd can mount up partitions
rmmod nbd
modprobe nbd max_part=8
RHEL 7: systemctl restart rapidrecovery-vdisk
RHEL 6: service rapidrecovery-vdisk restart
2. Perform a local mount again, remember to mount it as writable. If there are multiple disks to be mounted for lvm, please remember to mount all them up.
local_mount
Note: Not all the partitions would be mounted as a filesystem correctly. Some will be only present as devices
3. After mounting, to check the partition devices detected from nbd devices
lsblk -a
If these partitions contain filesystem rather than lvm, just mount it up manually if it is not mounted in local_mount
i.e. mount /dev/nbd0p1 /mnt2/boot
Step 4-5, 7 are LVM specific so you can ignore them if you are not using LVM
4. Import volume group in a new name (in the example: test) if the partitions contain LVM. This needs to be done usually as the volume group contains the same UUID as the existing ones already mounted, so needs to be imported in a different name. If there are multiple device for the volume group, just append all of them in the vgimportclone command.
vgimportclone -n test /dev/nbd0p2
vgchange -ay
5. Mount the lvm volume manually
mount /dev/mapper/test_lv1 /mnt2/lv1
6. After file browsing/copying from the mounted recovery points are complete, please unmount the mountpoints from the OS and then from local_mount
umount /mnt2/boot
local_mount
u /dev/nbd0
7. It may be noticed that the imported Volume Group (in the example: test) still exists but unable to be removed as the devices are all missing. It can be deleted using this workaround.
dd if=/dev/zero of=/tmp/tmp.raw bs=1M count=100
losetup -f
losetup /dev/loop0 /tmp/tmp.raw
vgextend test /dev/loop0
vgremove test -force
pvdevice pvremove /dev/loop0
© ALL RIGHTS RESERVED. Feedback 利用規約 プライバシー Cookie Preference Center