archinstall
archinstall copied to clipboard
archlinux
LVM partitions not deleted properly on wipe
Which ISO version are you using?
2024.12.01
The installation log
Could not determine the filesystem: None
Could not determine the filesystem: parted.FileSystem instance --
type: linux-swap(v1) geometry: <parted.geometry.Geometry object at 0x76c87589eba0> checked: False
PedFileSystem: <_ped.FileSystem object at 0x76c875876d80>
Could not determine the filesystem: None
https://0x0.st/XCM6.log or https://0x0.st/XCuP.log or https://0x0.st/XCub.log
describe the problem
Hardware model detected: Gigabyte Technology Co., Ltd. B650 GAMING X; UEFI mode: True Processor model detected: AMD Ryzen 5 7600X 6-Core Processor Memory statistics: 30701756 available out of 32463564 total installed Could not detect virtual system: ['/usr/bin/systemd-detect-virt'] exited with abnormal exit code [1]: none
System is not running in a VM: ['/usr/bin/systemd-detect-virt'] exited with abnormal exit code [1]: none
Virtualization detected: None; is VM: False Graphics devices detected: dict_keys(['Advanced Micro Devices, Inc. [AMD/ATI] Navi 22 [Radeon RX 6700/6700 XT/6750 XT / 6800M/6850M XT] (rev c1)'])
Creating partitions: /dev/nvme0n1
Type: primary
Filesystem: fat32
Geometry: 2048 start sector, 2097152 length
Type: primary
Filesystem: btrfs
Geometry: 2099200 start sector, 3904927920 length
Traceback (most recent call last):
File "/usr/lib/python3.12/site-packages/archinstall/__init__.py", line 337, in run_as_a_module
main()
File "/usr/lib/python3.12/site-packages/archinstall/__init__.py", line 330, in main
importlib.import_module(mod_name)
File "/usr/lib/python3.12/importlib/__init__.py", line 90, in import_module
return _bootstrap._gcd_import(name[level:], package, level)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "<frozen importlib._bootstrap>", line 1387, in _gcd_import
File "<frozen importlib._bootstrap>", line 1360, in _find_and_load
File "<frozen importlib._bootstrap>", line 1331, in _find_and_load_unlocked
File "<frozen importlib._bootstrap>", line 935, in _load_unlocked
File "<frozen importlib._bootstrap_external>", line 995, in exec_module
File "<frozen importlib._bootstrap>", line 488, in _call_with_frames_removed
File "/usr/lib/python3.12/site-packages/archinstall/scripts/guided.py", line 186, in <module>
guided()
File "/usr/lib/python3.12/site-packages/archinstall/scripts/guided.py", line 182, in guided
fs_handler.perform_filesystem_operations()
File "/usr/lib/python3.12/site-packages/archinstall/lib/disk/filesystem.py", line 70, in perform_filesystem_operations
device_handler.partition(mod, partition_table=partition_table)
File "/usr/lib/python3.12/site-packages/archinstall/lib/disk/device_handler.py", line 712, in partition
disk.commit()
File "/usr/lib/python3.12/site-packages/parted/decorators.py", line 28, in new
ret = fn(*args, **kwds)
^^^^^^^^^^^^^^^^^
File "/usr/lib/python3.12/site-packages/parted/disk.py", line 208, in commit
return self.__disk.commit()
^^^^^^^^^^^^^^^^^^^^
_ped.IOException: Partition(s) 2, 3 on /dev/nvme0n1 have been written, but we have been unable to inform the kernel of the change, probably because it/they are in use. As a result, the old partition(s) will remain in use. You should reboot now before making further changes.
It's the same error as https://github.com/archlinux/archinstall/issues/1789, the exception is still not caught.
Shouldn't it be caught and partprobe attempted, or user at least informed?
@C0rn3j
I disagree with using partprobe as a solution here. Can you identify what is causing the exception to occur?
The log shows that device nvme0n1 was being installed to and would be wiped, including of all existing partitions, by the installer. On that device, the existing partition nvme0n1p3 should have been wiped but was not (the partition is visible in the log under disk states of the second run after the first run should have wiped it). This partition therefore stays in use as the partition table is changed and thus the exception.
The following is pertinent information from the log about the partition:
{
"name": "nvme0n1p3",
"path": "/dev/nvme0n1p3",
"pkname": "nvme0n1",
...
"fstype": "LVM2_member",
"fsver": "LVM2 001",
...
"children": [
{
"name": "VG_XenStorage--13ab9ae7--5c42--a888--d71f--0e26ff49b0af-MGT",
"path": "/dev/mapper/VG_XenStorage--13ab9ae7--5c42--a888--d71f--0e26ff49b0af-MGT",
"pkname": "nvme0n1p3",
...
"type": "lvm",
...
},
{
"name": "VG_XenStorage--13ab9ae7--5c42--a888--d71f--0e26ff49b0af-VHD--759ca1c3--356e--41d9--890b--5f3e2dea82cf",
"path": "/dev/mapper/VG_XenStorage--13ab9ae7--5c42--a888--d71f--0e26ff49b0af-VHD--759ca1c3--356e--41d9--890b--5f3e2dea82cf",
"pkname": "nvme0n1p3",
...
"type": "lvm",
...
}
]
},
This is how device wiping is handled:
https://github.com/archlinux/archinstall/blob/d2ef961b3237b68147032e766b0ce512e9835d64/archinstall/lib/disk/device_handler.py#L824-L847
Apparently the device wiping logic is not sufficient for an LVM partition. This bug can be reproduced by installing to a device that contains an existing logical volume and having the device set to be wiped by the installer.
I hit this error and rebooting into the live install and rerunning resolved the issue.
I hit this error and rebooting into the live install and rerunning resolved the issue.
Can confirm.
On top of that I ran fdisk after reboot and completely wiped out all partitions on the device that had been created by archinstall. Not sure if it actually resolved anything but rerunning archinstall solved the issue.
