View Issue Details

Jump to NotesJump to History
IDProjectCategoryView StatusDate SubmittedLast Update
0017962CentOS-8anacondapublic2020-12-18 20:152020-12-18 20:15
Reporterps7776 Assigned To 
PrioritynormalSeverityminorReproducibilityalways
Status newResolutionopen 
Product Version8.2.2004 
Summary0017962: CentOS8.2 rescue image attempt at mounting system w/o valid bootstrap crashes rather than fail gracefully
DescriptionIf one boots the 8.2 rescue image to fix an existing system with say a missing grub2 bootstrap and selects option 2 ( search for installed systems ) in the last dialogue anaconda ( blivet? ) will crash and lock up. The only workaround as far as I can tell is to go for option 3, mount the partitions in question manually on /mnt/sysimage, mount --bind /dev, /sys ... manually and then do the chroot /mnt/sysimage. Then proceed to re-install grub2 .

 Rather than a hard crash if it cannot find the primary bootstrap (?) or the secondary grub2 code in the EF02 partition (?) the tool should preferably fail gracefully and inform the user that something is missing or mangled and perhaps still try to set up /mnt/sysimage in such a way that recovery is possible. I'm not sure exactly what the python code checks for. There may be other things in addition to the bootstrap missing that will trigger this bug. Being able to re-install the bootstrap to recover a damaged system is quite important

Steps To Reproduce Take an existing legacy install and make similar partitions on another otherwise blank drive. Create new filesystems with the same UUIDs on the new drive and rsync the original drive to the new ( offline ) . Create an EF02 bios boot partition on the new drive. Do not write to the EF02 partition.


This new disk will not have a valid bootstrap. To create one boot from a 8.2 rescue disk and select option 2 . Which result is a hard crash.


TagsNo tags attached.

Activities

There are no notes attached to this issue.

Issue History

Date Modified Username Field Change
2020-12-18 20:15 ps7776 New Issue