[qubes-users] Upgrading primary HD size

Mailing Lists qubes-users
1

Hi I recently upgraded to a new primary HD and these are the steps I’ve taken:

  1. plug the new HD in via USB

  2. boot from debian live

  3. use dd to copy my entire old HD to new HD

  4. use gdisk to convert from MBR to GPT

  5. use gparted to move the swap partition to the end of the drive, and resize the primary partition to use the remaining space

  6. swap in the new HD

I read I need to resize the LVM thin pool but, I’m not seeing the right output from lvs.

Existing threads:
https://groups.google.com/g/qubes-users/c/D-on-hSX1Dc/m/Q3rbYGyvAAAJ
https://groups.google.com/g/qubes-users/c/w9CIDaZ3Cc4/m/0xvtMUrIAgAJ

I also have a second 2TB drive with a second pool.

lsblk output:

nvme0n1 259:0 0 7.3T 0 disk
├─nvme0n1p3 259:3 0 15.4G 0 part
│ └─luks-[…] 253:1 0 15.4G 0 crypt [SWAP]
├─nvme0n1p1 259:1 0 1G 0 part /boot
└─nvme0n1p2 259:2 0 7.3T 0 part
└─luks-[…] 253:0 0 7.3T 0 crypt /
[…]

sda 8:0 0 1.8T 0 disk

└─luks-[…] 253:2 0 1.8T 0 crypt

├─qubes-poolhd0_tdata 253:4 0 1.8T 0 lvm

│ └─qubes-poolhd0-tpool 253:5 0 1.8T 0 lvm

[… my qubes on second HD]

└─qubes-poolhd0_tmeta 253:3 0 120M 0 lvm

└─qubes-poolhd0-tpool 253:5 0 1.8T 0 lvm

[… my qubes on second HD]

[…]

$ qvm-pool -l
NAME DRIVER
varlibqubes file-reflink
linux-kernel linux-kernel
poolhd0_qubes lvm_thin

$ sudo lvs -a
LV VG Attr LSize Pool Origin Data% Meta% Move Log Cpy%Sync Convert
[lvol0_pmspare] qubes ewi------- 120.00m
poolhd0 qubes twi-aotz-- 1.82t 69.41 43.01
[poolhd0_tdata] qubes Twi-ao---- 1.82t
[poolhd0_tmeta] qubes ewi-ao---- 120.00m
[… my qubes on second HD]

Where have my Qubes on the first HD gone? They still work, but I don’t see them in the output of these commands.

2

Oops, I forgot I’m using btrfs.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐

3

For the record it’s:

sudo btrfs filesystem resize +1t /

sudo btrfs filesystem resize +100g /

sudo btrfs filesystem resize +10g /

etc

Just keep spamming until it gives you errors.

‐‐‐‐‐‐‐ Original Message ‐‐‐‐‐‐‐

4

Oops, I forgot I'm using btrfs.

Well, it's not specific to Qubes OS, but maybe you'd like to read this:
Setting up a HA cluster using Xen PVMs recently, I found a bug that activated one VM on two nodes at the same time... The VM was using BtrFS as root / boot filesystem with many subvolumes and automatic snapshots before each software update.
As a result the BtrFS was corrupted, and there was NO way to recover any of the snapshots or subvolumes. Maybe keep this in mind. In the past I'd traditionally use separate ext2/3 filesystems for things like /, /boot, /var, etc. And the changes to recover something are probably higher than with BtrFS... Anyway: I just wanted to tell.

Regards,
Ulrich