Restoring a Volume to Software-RAID1

[MyKey0815]

I have a Univention Coorparate Server (UCS) installed with Raid1. On Weekend the machine is going brocken through missconfiguration so i will restore the complete Filedata to the same system.

My first attemp was to restore the LVM-Data to /dev/sda3 (raid) and /dev/sdb3 (raid)

After reboot the problem exists.

Now I try to restore the LVM only to /dev/sda3 - but now it will not be shown anymore as a part of a raid.

What is the correct way to restore a Volumen to software RAID?

[nielsengelen]

To be correct: you used the bare metal recovery iso to perform this restore?

Which error are you getting after the reboot? Is this GRUB related or anything else? Did you by any chance contact support already?

[MyKey0815]

I use veeam-recovery-media-1.0.0.944_x86_64.iso for restoring

I seen no error. Only "Please wait, loading ...." and nothing happens. I think, that my Raid is totaly brocken. I will recreate the raid configuration and then try another restore. Can you please descripe what I must chose (source and destinantion)?

https://ibb.co/h9kmak

[MyKey0815]

Here some details about my configuration:

The OS is "Debian 7" with mdadm as software RAID. I have fresh installed the UCS-Server with all the updates (there are some Kernelupdates in the meanwhile).

Now I want to restore my backup. What are the right way to restore only the LVM-Volume when under "Current System" i see 2 Harddisks with each 3 Partions (marked as (raid))?

[MyKey0815]

So, today I will recover some files from my veeam-image to the raid-server. But the problem is, that they cant mount the "current system". Backup-Share is succesfull mountet, but "/mnt/system" is empty.

I will now wait until the raid is synced successfull. Then I will restart the restore process. But can someone help me how to restore the file/volumes to this server?

[nielsengelen]

If the backup is mounted it should be under /mnt/backup normally. However if you don't see any files there I suggest you contact support and maybe a remote session to see where something might be going wrong so you can recover the files.

[MyKey0815]

The problem is NOT the /mnt/backup. There I can see all the files of the backup.

My problem ist /mnt/system. This directory is empty. I think, that veeam-linux agent restore is not able to see the mdadm Raid1 correctly or can not mount them.

[PTide]

Hi,

You need to drop into command line in recovery media and use built-in mdadm to activate the array. If that does not for work for some reason then please contact our support team directly.

Thanks

[MyKey0815]

Sorry - i start with veeam-recovery-media-1.0.0.944_x86_64.iso from CD and select "Switch to commandline".

Then I type in "mdadm" and the error came up: "mdadm: command not found"

So i didn´t can activate the raid. But what are the commands to activate them? mdadm is the tool. But i think i must type in some options. Or what you mean?

[MyKey0815]

Today i try the new restore-media-iso. But the problems still exist

[CptLuxx]

Hi,

i made a Backup with the latest veeam agent free version from an debian 8 Server with an mdam raid.. Unfortunately im unable to restore the Backup

maybe some can tell me how i should restore it? Here is an screenshot from the recovery cd

[PTide]

Hi,

Currently there is no mdadm support in recovery media, though it's planned. You can contact our support team to get instructions on how to workaround this problem using third-party liveCD (basically any distro LiveCD of your choice tha contains mdadm.)

Thanks

[sidse]

Hi PTide
Is there an instruction, how a veeam backup could manually restored by a linux system?
We have the same situation with a SWRaid and want to see a way of recover backups.

Thanks
Roman

[PTide]

Hi,

The easiest way would be to use any distro's liveCD with mdadm utility in it, and install the agent there. RAID tools will be added to veeam recovery media soon.

Thank you

[zlydk]

Hi,

I recently had to restore a remote (out of band) system from bare metal.
I was challenged by the fact that the Recovery Media does not currently contain the tools needed to assemble md raid volumes (mdadm), prior to restoring.
Eventually I used a livecd distro, which I was able to boot remotely via IDE-R, install the Veeam Agent in the live environment, import the backup, and perform the restore from shell using the veeam commandline interface.

It would have made the restore process a whole lot easier (for me), if the Recovery Media contained the raid tools needed to assemble the arrays.

I saw a couple of references on the forum already, which states that mdadm is planned to be included in the recovery media (e.g. post260584.html#p274352)

Any ETA on that?

[PTide]

Hi,

It will be added in the upcoming release.

Thanks

[MyKey0815]

Hello again,

i have seen, that the software-raid is now included in the restore-cd (v2.0).

Can someone describe how i can restore the full machine backup to a new harddrive (same size as the original).

My Problem: I have a mdadm-Raid on my machine and i will find out, how i can do a downgrade/remove the mdadm for easier backup with veeam.

If i start the recovery cd then i see on the left side:

Code: Select all

sda - 1.82TB

On the right side the structure of the Raid:

Code: Select all

md0 952.4M /boot (ext2)
sda (boot) 1.82T
sbb (boot) 1.82T
vg_ucs_root 1.76TB
  lv_ucs_root 1.76T / (ext4)

Can anybody describe the needed steps for restore to ONE Harddrive?

[PTide]

Hi,

Where does that vg_ucs_root reside, is it on RAID too? What is the output of your lsblk -af?

In general, you need to select that "/boot" on the right and restore it to sda on the left. That will automatically create partition of a required size. Next, you select lv_ucs_root logical volume on the right and restore it to the remaining FREE space on the left. Or, alternatively, you can select the remaining FREE space and create a VG out of it, create a logical volume and restore the lv_ucs_root there.

Thanks!

[MyKey0815]

Thank you for the reply. Here the result:

Code: Select all

root@ucs100091:~# lsblk -af
NAME                          FSTYPE            LABEL                 UUID                                   MOUNTPOINT
loop0
loop1
loop2
loop3
loop4
loop5
loop6
loop7
sda
├─sda1                        linux_raid_member unassigned-hostname:0 8ca29786-b4d2-6db4-fbcb-8f629636b792
│ └─md0                       ext2                                    85baf293-8fab-40bf-99a8-3b69f17b622a   /boot
├─sda2                        linux_raid_member unassigned-hostname:1 baa72337-41f9-9d7c-39af-4fdf721e4aad
│ └─md1                       swap                                    f35aff87-d944-4013-9222-411010d7c21f   [SWAP]
└─sda3                        linux_raid_member unassigned-hostname:2 2013bcbf-62ba-02ca-ca8e-9a1f5d133de5
  └─md2                       LVM2_member                             xl7Rx3-20MM-tOAP-e9ep-xgiU-gawr-Sc17Bd
    └─vg_ucs_root-lv_ucs_root ext4                                    61db28f8-9421-4638-a7e6-700bec3a1eeb   /
sdb
├─sdb1                        linux_raid_member unassigned-hostname:0 8ca29786-b4d2-6db4-fbcb-8f629636b792
│ └─md0                       ext2                                    85baf293-8fab-40bf-99a8-3b69f17b622a   /boot
├─sdb2                        linux_raid_member unassigned-hostname:1 baa72337-41f9-9d7c-39af-4fdf721e4aad
│ └─md1                       swap                                    f35aff87-d944-4013-9222-411010d7c21f   [SWAP]
└─sdb3                        linux_raid_member unassigned-hostname:2 2013bcbf-62ba-02ca-ca8e-9a1f5d133de5
  └─md2                       LVM2_member                             xl7Rx3-20MM-tOAP-e9ep-xgiU-gawr-Sc17Bd
    └─vg_ucs_root-lv_ucs_root ext4                                    61db28f8-9421-4638-a7e6-700bec3a1eeb   /
sr0

I have try your mention steps. But when i restore "/boot" then i have no "Free space" to restore "lv_ucs_root". Which additional steps are needed?

[PTide]

It seems that automatic creation didn't work. Then please select "sda" on the left, press Enter, create partition table, and after that you should see "free" under "sda".
Press 'Enter' on that "free" entry, choose "restore volume from..." and select "/boot" on the right. Repeat the procedure for "/" partition.

Thanks!

[MyKey0815]

Thank you. In the last days i have do multiple Restores of the partions. But "the best" result is follwing error:

Code: Select all

Busybox 1.22 .......

(initramfs)

I didn´t find a working way to reaktivate the backup on the new drive.

I have try to start with the rescue-disc and repair the grub. But that is not enough.

[PTide]

It might be so that your filesystem needs to be checked. Try to type exit while in the initramfs commandline and see if it gives you a hints on which filesystem is not ok. After that run fsck against the device.

Thanks!

[MyKey0815]

If i execute exit then come up following hint:

Code: Select all

ALERT! /dev/mapper/vg_ucs_root-lv_ucs_root does not exists. Dropping to a shell

[MyKey0815]

Now I will try following steps:

1.) Recreate the same partioning of the new disc with the paramters of the source disk
- sda1 1GB
- sda2 66GB
- sda3 1.9TB
2.) Start the veeam-recovery cd and connect to shared folder with image
3.) Select current image and snapshot
4.) Restore md0 to sda1
5.) On sda3 create a LVM
- Groupname is the same like the source: vg_ucs_root
6.) Now a new free partion is under the vg_.... Select this and restore lv_ucs_root to this partition

I hope that it works - in 5 hours i will it know

[PTide]

Right, the error means that you need to either

a) modify grub config so that it will search for the new root partition instead of the old root logical volume
b) create volume group and a logical volume with the same name during restore

Thanks

[MyKey0815]

Here are the missing steps after restore

7.) Download Super GRUB2 Disk http://www.supergrubdisk.org/super-grub2-disk/
- the menu entry “Enable GRUB2’s RAID and LVM support” has to be selected. Activating the RAID/LVM support may take a moment. This step may be omitted and can be catched up on later if neccessary.
- Select the entry “Detect and show boot methods”. Please note: depending on the complexity of your setup, this might take a moment. A black screen is shown while detecting.
- The next menu shows all available boot options. Select an appropriate entry containing the text “Univention Corporate Server”.
8.) Repair GRUB
- login UCS
- Open Terminal/Console
- sudo su
dpkg-reconfigure grub-pc
“Linux command line”: Press Return to keep the value as-is
“default command line for Linux”: Press Return to keep the value as-is
Question for changed configuration file „/etc/default/grub“: Press Return to keep the configuration file as-is.
“devices for GRUB installation”: select the correct device(s) (i.e. the device(s) the BIOS is configured to boot from) to install GRUB on.
Multiple devices can be selected.
Please note: Select only devices (e.g. /dev/sda), not partitions (e.g. /dev/sda1). Do not select device mapper devices (e.g. /dev/dm-0).
ucr commit /etc/default/grub
9.) Reboot


Thanks for all help here. Now I can use the restore!!!!!

[PTide]

Thank you for getting back to us with that. Maybe we could implement some sort of semi-automatic post-restore routines for advanced restore scenarios (migration from/to RAID, dissimilar hardware restores, etc.) in the next versions of recovery ISO.

Thanks!