Join GitHub today
GitHub is home to over 28 million developers working together to host and review code, manage projects, and build software together.
Sign upInstaller problem when deleting some partitions but keeping one #3843
Comments
andrewdavidwong
added
bug
C: installer
labels
Apr 22, 2018
andrewdavidwong
added this to the Release 4.0 updates milestone
Apr 22, 2018
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
trueriver commentedApr 22, 2018
Qubes OS version:
Affected component(s):
R4.0 installer (only tested from USB stick)
Steps to reproduce the behavior:
find a computer with a single disk whose contents you are happy to trash.
Using the live Linux OS of your choice unmount any partitions that the OS mounted, open fdisk, delete all partitions and create three new ones. I created the first with +1024 sectors, the next with +100G space, and the third to fill the rest of the disk.
Exit from fdisk. Reboot if you get a message saying that the kernel has not been updated with the new partitions.
Still (or back) in the live OS, create ext4 filesystems in each partition.
Shutdown. Remove the Live OS boot media, and insert the Qubes installer media. Boot into the R4 installer.
Select language
On the next page, go into "Installation Destination"
Select "Automatically configure partitioning" and "I would like to make additional space available"
In my tests I also unselected encryption (but I am guessing this makes little difference)
Click "Done" top left.
Expected behavior:
Partitioner appears
Actual behavior:
Diaslog box "An unknown error has occurred"
The further details begin with the foillowing
anaconda 25.20.9-12 exception report
Traceback (most recent call first):
File /usr/lib64/python3.5/site-pachakes/parted/alignment.py in alignUp
I am copying this by hansd from one machione to another so will not copy any more.
General notes:
Related issues:
#3842
I expected this to produce the same result as #3842 but am posting it separately because this causes Anaconda to crash rather than reporting a fatal error
My initial intention was to find the simplest way to trigger that bug, but instead I have either found a different manifestation of it, or a different bug entirely.
Because of the Anaconda crash I am wondering if this is an upstream problem?