2

On my Debian 8 server HDD /dev/sda crashed. mdadm informed me via email and I had the disk replaced.

After the server was back up I copied over my GPT from using sgdisk -R /dev/sdb /dev/sda. The second I hit "Enter" on my keyboard I realized my mistake.

So now I have an empty GPT on both disks.

My question is if it is possible to re-create the GPT on /dev/sdb as the server is still running as I did not reboot since copying the wrong GPT?

I did a backup with sfdisk -d /dev/sdb > sdb.partition.table before the faulty HDD was replaced. But as I did not do a backup with sgdisk the backup is completely useless, if I am correct?

Additionally I have this output from fdisk -l from before copying the GPT:

Disk /dev/sdb: 2.7 TiB, 3000592982016 bytes, 5860533168 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes Disklabel type: gpt Disk identifier: 454774BD-960F-45C6-8C82-AE5C156444E0 Device Start End Sectors Size Type /dev/sdb1 4096 33558527 33554432 16G Linux RAID /dev/sdb2 33558528 34607103 1048576 512M Linux RAID /dev/sdb3 34607104 5860533134 5825926031 2.7T Linux RAID /dev/sdb4 2048 4095 2048 1M BIOS boot Partition table entries are not in disk order. Disk /dev/md0: 16 GiB, 17171349504 bytes, 33537792 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes Disk /dev/md1: 511.7 MiB, 536543232 bytes, 1047936 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes Disk /dev/md2: 2.7 TiB, 2982739705856 bytes, 5825663488 sectors Units: sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 4096 bytes I/O size (minimum/optimal): 4096 bytes / 4096 bytes
Improve this question

1 Answer 1

Reset to default
0

After researching a bit and after trying out tools such as testdisk, I did not find a defintive way to restore my original GPT on /dev/sdb

So I tried using cgdisk and it was successfull, as I still had the original "sector layout" of /dev/sdb noted down:

Disk identifier: 9F95A04D-3ECB-144D-B2A0-55CDD986072B Device Start End Sectors Size Type /dev/sdb1 4096 33558527 33554432 16G Linux RAID /dev/sdb2 33558528 34607103 1048576 512M Linux RAID /dev/sdb3 34607104 5860533134 5825926031 2.7T Linux RAID /dev/sdb4 2048 4095 2048 1M BIOS boot

With this information I created partitions with the same starting sectors, the same number of sectors used and the same file system types as stated above.

After writing the GPT on /dev/sdb, fdisk -l /dev/sdb gave me the same output as above.

I then copied over the GPT by using sgdisk -R /dev/sda /dev/sdb (this time in the correct order) and fdisk -l /dev/sda showed me the exact same "sector layout" for /dev/sda as for /dev/sdb:

Disk identifier: 4CB38488-8B72-44AA-8449-4E4692165893 Device Start End Sectors Size Type /dev/sdb1 4096 33558527 33554432 16G Linux RAID /dev/sdb2 33558528 34607103 1048576 512M Linux RAID /dev/sdb3 34607104 5860533134 5825926031 2.7T Linux RAID /dev/sdb4 2048 4095 2048 1M BIOS boot


All that was left to do, was to resync the RAID-Volumes using mdadm and re-install grub2.

After resyncing was done, as mentioned, I re-installed grub2 on /dev/sda (I re-installed it /dev/sdb too, just to be sure) and generated a new devicemap. (I had to flush the HDD buffers to avoid the grub2 core image warnings, though)

I rebooted the server and it came up again just fine.

IMPORTANT: I only did this GPT tinkering because I have a complete backup of my server as I was not 100% sure this would work and I could have destroyed my partitions.

Improve this answer

You must log in to answer this question.

Start asking to get answers

Find the answer to your question by asking.

Ask question

Explore related questions

See similar questions with these tags.