Tails Installer should not run update_system_partition_properties on MBR partitions
From WhisperBack report 4d062594b54740b1f53a3a64c4c7032d.
While doing an "Upgrade from ISO", someone got the following trace:
tails-i386-1.1.iso selected Unmounting mounted filesystems on '/dev/sdd1' [...] Updating properties of system partition /dev/sdd1 Error modifying partition: helper exited with exit code 1: In part_change_partition: device_file=/dev/sdd, start=31744, new_start=31744, new_size=3929335808, type=XXXXXXXX-XXXX-XXXX-XXXX-XXXXXXXXXXXXX Entering MS-DOS parser (offset=0, size=3930062848) MSDOS_MAGIC found looking at part 0 (offset 31744, size 3929335808, type 0x0c) new part entry looking at part 1 (offset 0, size 0, type 0x00) new part entry looking at part 2 (offset 0, size 0, type 0x00) new part entry looking at part 3 (offset 0, size 0, type 0x00) new part entry Exiting MS-DOS parser MSDOS partition table detected containing partition table scheme = 0 invalid type 'XXXXXXXX-XXXX-XXXX-XXXXXXXXXXXX' given
It seems like they have a MBR partition table, which means that the device was not installed with Tails Installer, but most likely with the UUI. It seems that Tails Installer (when doing any kind of upgrade) should check what partition table the destination device has, and only run
update_system_partition_properties() if that's a GPT.
The code lives in https://git-tails.immerda.ch/liveusb-creator. We already have helpers to talk to udisks there, and likely the needed info is already available in some object attribute.
That's a regression in 1.1, maybe not a blocker for 1.1.1.
#1 Updated by intrigeri over 4 years ago
- Description updated (diff)
- Assignee set to kurono
- Starter set to Yes
(Improved description => flagged as Easy.)
kurono, do you want to give it a try? If you do, please adjust the target version field according to your ETA. Otherwise, just deassign this ticket from you. Thanks!
#4 Updated by intrigeri over 4 years ago
- % Done changed from 0 to 20
Thanks! kurono, when flagging a ticket as ready for QA, please deassign it from you, so that it's clear that nobody volunteered yet to take care of the review. The process is documented on https://tails.boum.org/contribute/merge_policy/#index2h1.