Raid 5 restore after OMV reinstall

Hikari · Mar 25th 2020

Hello guys, I would need some help to rebuild my RAID 5 and recover my data.

My config :

- HP Proliant Gen8

- ESXI on SD card

- Virtual machine with OMV on SSD

- 4x 3To WD red in raid 5. Disks are directly mapped to OMV.

What's happen:

ESXI crash, I had to reinstall it. I also lost VM OMV and had to reinstall it.

I have mapped again the disks to OMV but from what I have understood I have to rebuild the RAID.

Looking to those log it seems 1 disk is not ok..

Code

root@openmediavault:~# cat /proc/mdstat
Personalities : [linear] [multipath] [raid0] [raid1] [raid6] [raid5] [raid4] [raid10]
md127 : inactive sdd[2](S) sde[0](S) sdb[1](S)
      8790308232 blocks super 1.2

unused devices: <none>

Code

root@openmediavault:~# blkid
/dev/sda1: UUID="e2f43903-97ee-40cf-8071-b71c7925c637" TYPE="ext4" PARTUUID="c684c0c4-01"
/dev/sda5: UUID="8b1d20b7-dad5-4c62-8dee-a1db7a32d4ee" TYPE="swap" PARTUUID="c684c0c4-05"
/dev/sdb: UUID="c91b4ded-2fd5-9908-936a-1f3402af4b25" UUID_SUB="39beb9de-b9bd-71f1-3c68-4f9a50d53d86" LABEL="omv:OMV" TYPE="linux_raid_member"
/dev/sde: UUID="c91b4ded-2fd5-9908-936a-1f3402af4b25" UUID_SUB="a7357e6c-b9de-4b58-a8a4-f9463ae8fb74" LABEL="omv:OMV" TYPE="linux_raid_member"
/dev/sdd: UUID="c91b4ded-2fd5-9908-936a-1f3402af4b25" UUID_SUB="ee554dab-d234-64a2-8757-ee37af016982" LABEL="omv:OMV" TYPE="linux_raid_member"
/dev/sdc: PTUUID="edf65d70-51df-4bf4-9b5a-559e18a68199" PTTYPE="gpt"

Code

root@openmediavault:~# fdisk -l | grep "Disk "
Partition 1 does not start on physical sector boundary.Disk /dev/sda: 16 GiB, 17179869184 bytes, 33554432 sectors
Disk model: Virtual disk
Disk identifier: 0xc684c0c4
Disk /dev/sdb: 2.7 TiB, 3000559427584 bytes, 5860467632 sectors
Disk model: LOGICAL VOLUME
Disk identifier: 0xa358c847

Partition 1 does not start on physical sector boundary.Disk /dev/sde: 2.7 TiB, 3000559427584 bytes, 5860467632 sectors
Disk model: LOGICAL VOLUME
Disk identifier: 0x379c8e05

Partition 1 does not start on physical sector boundary.Disk /dev/sdd: 2.7 TiB, 3000559427584 bytes, 5860467632 sectors
Disk model: LOGICAL VOLUME
Disk identifier: 0x000557d7

Disk /dev/sdc: 2.7 TiB, 3000559427584 bytes, 5860467632 sectors
Disk model: LOGICAL VOLUME
Disk identifier: EDF65D70-51DF-4BF4-9B5A-559E18A68199

Display More

Code

root@openmediavault:~# cat /etc/mdadm/mdadm.conf
# mdadm.conf
#
# !NB! Run update-initramfs -u after updating this file.
# !NB! This will ensure that initramfs has an uptodate copy.
#
# Please refer to mdadm.conf(5) for information about this file.
#

# by default (built-in), scan all partitions (/proc/partitions) and all
# containers for MD superblocks. alternatively, specify devices to scan, using
# wildcards if desired.
#DEVICE partitions containers

# automatically tag new arrays as belonging to the local system
HOMEHOST <system>

# instruct the monitoring daemon where to send mail alerts
MAILADDR root

# definitions of existing MD arrays

# This configuration was auto-generated on Mon, 24 Feb 2020 19:58:45 +0000 by mkconf

Display More

Code

root@openmediavault:~# mdadm --detail --scan --verbose
INACTIVE-ARRAY /dev/md127 num-devices=3 metadata=1.2 name=omv:OMV UUID=c91b4ded:2fd59908:936a1f34:02af4b25
   devices=/dev/sdb,/dev/sdd,/dev/sde

Thanks for your support !

geaves · Mar 26th 2020

I have no idea how ESXI behaves having checked a VM of OMV running on VB the output is different.

Anyway, the array is inactive as the output suggests from mdsat and mdadm --detail, however, that output shows 3 drives /dev/sd[bde] and that is confirmed from the output of blkid. There is also no definition in mdadm conf.

I assume that /dev/sdc should also be part of the array, (you state in your first post there are 4 drives) but that drive does not have a "linux raid member" signature.

When you reinstalled ESXI and subsequently OMV were those 4 drives disconnected? do have a backup?

Normal procedure to get an inactive array back up;

mdadm --stop /dev/md127

mdadm --assemble --force --verbose /dev/md127 /dev/sd[bde]

I have no idea if that will work or if your data will be accessible.

Hikari · Mar 26th 2020

Hello geaves,

I have a backup, but not with everything, so I would like to access the files in order to retrieve the missing ones if possible.

I don't understand why 1 of the 4 disk behaves differently than the others and is not pointed as a "linux_raid_member".

Anyway, if I reactive the array on 3 disks only, will I have access to the data ?

My goal next is to delete the RAID and go for another configuration (snapraid or other, I am still not sure)

Thanks for your support!

geaves · Mar 26th 2020

Quote from Hikari

Anyway, if I reactive the array on 3 disks only, will I have access to the data ?

TBH I have no idea, reactivating a raid is not normally an issue it's as if yours has had a drive physically removed i.e. as you would a hot swap except mdadm does not support hot swap.

A Raid 5 will allow for one drive failure, will the re assembly fail, technically it shouldn't, then you have lost your data.

Hikari · Mar 26th 2020

Hello geaves,

I applied the procedure to restore the array, and it works with the 3 disks.

State of the array is clean, degraded (as expected) but I was able to create a system file and to access my data.

Now I just have to find the best backup method to replace RAID5 .

Thanks again for your support

geaves · Mar 26th 2020

Quote from Hikari

Thanks again for your support

You're welcome, pleased it worked

Raid 5 restore after OMV reinstall

Hikari Mar 25th 2020

Participate now!

Tags