Hey mikeaa, thanks for chiming in! I'm not sure I agree there actually, by looking at the current code. We only get to the point where grub-probe exits with a non-zero status due to grub-mount succeeding before (in the previous if condition). This is also the original report of the bug, as I understood it: filesystems are left mounted due to grub-mount succeeding, but grub-probe exploding the script prematurely (due to set -e and non-zero exit status).
Given the other 50mounted-tests script under os-probes/, I think we are meant to run tests if grub-mount succeeds, regardless of whether grub-probe fails (or returns an empty string). That's already the case for grub-probe exiting with zero but not printing out anything, which we currently address by forcing $type=fuseblk. My understanding is that if grub-mount succeeds, we do want to run the tests so that we can pick up any bootable OSes (i.e. just because we let the tests run, doesn't mean that a GRUB entry is going to be created; one of the OS tests needs to succeed for that). What do you think? -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/1987679 Title: os-prober leaves filesystems (lvm-thin, lvm snap) mounted To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/os-prober/+bug/1987679/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
