Ubuntu server 22.04 installation crash with intel VROC sata raid

Bug #1960392 reported by shangsong
6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
subiquity
New
Undecided
Unassigned
linux (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Reproduce steps:
1.Setup raid 0 with 2 sata disks and intel VROC SATA controller.
2.Install Ubuntu server 22.04 with the latest image.
3.It crash with error "devices or container for raid must be specified"

Revision history for this message
shangsong (shangsong2) wrote :
Revision history for this message
Ubuntu Kernel Bot (ubuntu-kernel-bot) wrote : Missing required logs.

This bug is missing log files that will aid in diagnosing the problem. While running an Ubuntu kernel (not a mainline or third-party kernel) please enter the following command in a terminal window:

apport-collect 1960392

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

Changed in linux (Ubuntu):
status: New → Incomplete
Revision history for this message
shangsong (shangsong2) wrote :

Fail to run the command "apport-collect 1960392".
Please check if the sosreport of #1 is enough, thanks.

Revision history for this message
Mauricio Faria de Oliveira (mfo) wrote :

The issue seems to be in subiquity (server installer) not the linux kernel, per the 'error' screenshot (thanks!)

Marking the subiquity project as affected, and linux as invalid.

Changed in linux (Ubuntu):
status: Incomplete → Invalid
Revision history for this message
Michael Hudson-Doyle (mwhudson) wrote :

So the reason this doesn't work is that the udev data for the container, /dev/md/imsm0, does not contain the keys we look for to work out which devices are part of it. We expect to find keys called things like MD_DEVICE_dev_sda_DEV but they are not there. There is UDISKS_MD_DEVICE_dev_sda_DEV though and this is very strange because the code to produce the MD_DEVICE and UDISKS_MD_DEVICE keys is essentially the same:

IMPORT{program}="BINDIR/mdadm --detail --export $devnode"

SUBSYSTEM=="block", KERNEL=="md*", ENV{DEVTYPE}!="partition", IMPORT{program}="/bin/sh -c '/sbin/mdadm --detail --export $tempnode | /bin/sed s/^MD_/UDISKS_MD_/g'"

I don't know why one use tempnode and one devnode -- maybe that's significant? Or maybe it's a race where the mdadm rules run before the devices is really ready or something. Does this happen every time?

Revision history for this message
shangsong (shangsong2) wrote :

 Does this happen every time?
-->Yes, recurrence rate 100%

Revision history for this message
Jeff Lane  (bladernr) wrote :

This is a duplicate of the linked bug that was filed by the upstream.

Revision history for this message
shangsong (shangsong2) wrote :

It can pass with the latest edge subiquity

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.