[linux-lvm] replacing missing disk thin pool raid1

[Etienne Champetier]

Hi All,

I'm doing some experiments before using lvm raid features, and I'm
stuck on a basic use case, replacing a disk :D

All my tests are done using an ubuntu 17.10 server vm
# lvs --version
  LVM version:     2.02.168(2) (2016-11-30)
  Library version: 1.02.137 (2016-11-30)
  Driver version:  4.37.0
# uname -a
Linux ubuntu 4.13.0-21-generic #24-Ubuntu SMP Mon Dec 18 17:29:16 UTC
2017 x86_64 x86_64 x86_64 GNU/Linux

If I just have a normal raid1 lv, for exemple my root
# lvconvert -m1 ubuntu-vg/root

if I wipe 1 of the disks and reboot (simulating a dead disk)
# dd if=/dev/zero of=/dev/vdb status=progress
# reboot

My recovery procedure would be:
# sfdisk -d /dev/vda | sfdisk /dev/vdb
# pvcreate /dev/vdb1
# vgextend ubuntu-vg /dev/vdb1
# lvconvert --repair ubuntu-vg/root
# vgreduce --removemissing ubuntu-vg


Now if I want to use lvm thin pool + raid1

# lvcreate -m1 --type raid1 -L1G ubuntu-vg -n thin_pool
# lvcreate -m1 --type raid1 -L8MB ubuntu-vg -n thin_meta
# lvconvert --thinpool ubuntu-vg/thin_pool --poolmetadata ubuntu-vg/thin_meta
# lvcreate -T ubuntu-vg/thin_pool -V300M -n thinlv

If I redo my recovery procedure, trying repair the thin_pool give me
# lvconvert --repair ubuntu-vg/thin_pool
  WARNING: Device for PV XiQsNI-J6CS-rpQa-yX1H-DA43-13SL-fxeSlf not
found or rejected by a filter.
  Using default stripesize 64,00 KiB.
  WARNING: recovery of pools without pool metadata spare LV is not automated.
  WARNING: If everything works, remove ubuntu-vg/thin_pool_meta0 volume.
  WARNING: Use pvmove command to move ubuntu-vg/thin_pool_tmeta on the
best fitting PV.

I've tried random commands but I'm stuck

My questions:
1) is my first recovery procedure "best practice" ?
2) what is the procedure to recover my thin pool + raid1 ?
3) Is thin pool + raid1 too new / should I stick to dmraid raid1 + lvm on top ?

Thanks in advance
Etienne