[VROC] Installation is available on spare disk, but it fails

Bug #1992459 reported by Kinga Tanska

This bug report was marked for expiration 218 days ago. (find out why)

6
This bug affects 1 person
Affects Status Importance Assigned to Milestone
subiquity (Ubuntu)
Incomplete
Undecided
Unassigned

Bug Description

Found on Ubuntu 22.04.1

Steps to reproduce:
1. Create container in OS with one disk (to mark its state as spare)
#mdadm --create /dev/md/imsm0 --metadata=imsm --raid-devices=1 /dev/nvmeXn1 –run
2. Verify that target disk is marked as spare. Get its metadata by executing command:
#mdadm --examine /dev/nvmeXn1
and check its state in output. Expected data:
State : spare
3. Reboot to target OS installer.
Note: Use VMD replacement driver if exists in tested build.
4. Check if spare disk cannot be chosen as installation destination

Expected result:
Spare disk cannot be chosen as installation destination.

Actual result:
Spare disk can be chosen as installation destination.

Additional info:
Installation on chosen spare disk can be started, but error occurs few seconds after installation starts.

Install log contains an error:
mdadm: successfully stopped /dev/md127 after 1 attempt(s)
        Wiping mdadm member devices: ['/dev/nvme3n1']
        Running command ['mdadm', '--examine', '/dev/nvme3n1'] with allowed return codes [0] (capture=True)
        finish: cmd-install/stage-partitioning/builtin/cmd-block-meta/clear-holders: FAIL: removing previous storage devices
        TIMED BLOCK_META: 1.765
        finish: cmd-install/stage-partitioning/builtin/cmd-block-meta: FAIL: curtin command block-meta
        Traceback (most recent call last):
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/commands/main.py", line 202, in main
            ret = args.func(args)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/log.py", line 97, in wrapper
            return log_time("TIMED %s: " % msg, func, *args, **kwargs)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/log.py", line 79, in log_time
            return func(*args, **kwargs)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/commands/block_meta.py", line 108, in block_meta
            meta_clear(devices, state.get('report_stack_prefix', ''))
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/commands/block_meta.py", line 1984, in meta_clear
            clear_holders.clear_holders(devices)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/block/clear_holders.py", line 638, in clear_holders
            shutdown_function(dev_info['device'])
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/block/clear_holders.py", line 209, in shutdown_mdadm
            mdadm.zero_device(mddev, force=True)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/block/mdadm.py", line 394, in zero_device
            metadata = mdadm_examine(devpath, export=False)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/block/mdadm.py", line 254, in mdadm_examine
            data = __mdadm_detail_to_dict(out)
          File "/snap/subiquity/3698/lib/python3.8/site-packages/curtin/block/mdadm.py", line 630, in __mdadm_detail_to_dict
            raise ValueError('Duplicate key in mdadm regex parsing: ' + key)
        ValueError: Duplicate key in mdadm regex parsing: state
        Duplicate key in mdadm regex parsing: state

Revision history for this message
Kinga Tanska (ktanska) wrote :
Revision history for this message
Dan Bungert (dbungert) wrote :

We fixed something similar sounding, would you retest with the beta version?
In the installation environment, you can get to a shell from the Help Menu -> Enter shell. At that terminal, would you do:

sudo snap refresh --beta subiquity

and see if things get better?

Revision history for this message
Lukas Märdian (slyon) wrote :

Setting status to "Incomplete", while we're waiting for this test to be confirmed.

Changed in subiquity (Ubuntu):
status: New → Incomplete
Revision history for this message
Kinga Tanska (ktanska) wrote :

Hi,

defect occurs also with beta version.

Regards,
Kinga Tanska

Revision history for this message
Kinga Tanska (ktanska) wrote :

Due to closing our project in Intel, there is no reason to process it.
Please close this issue.

Revision history for this message
Launchpad Janitor (janitor) wrote :

[Expired for subiquity (Ubuntu) because there has been no activity for 60 days.]

Changed in subiquity (Ubuntu):
status: Incomplete → Expired
Revision history for this message
Kinga Tanska (ktanska) wrote :

Hi,

I've changed status to New, because issue still occurs.
We found similar behavior on UB 20.04.3 too. Spare is available as installation destination, and installation starts, but error occurs few second after start.

I attached installer logs for this case.

I will be very thankful if you will look at this.

Regards,
Kinga

Changed in subiquity (Ubuntu):
status: Expired → New
Revision history for this message
Kinga Tanska (ktanska) wrote :
Revision history for this message
Dan Bungert (dbungert) wrote :

Hi Kinga,

Many raid related changes have happened since the build you have tested, and in fact we landed another fix recently, please retest with the latest version. You can do this with a supported install ISO or Ubuntu Mantic daily, then my previous guidance still applies:

In the installation environment, you can get to a shell from the Help Menu -> Enter shell. At that terminal, would you do:

sudo snap refresh --beta subiquity

and see if things get better?

Note that, I don't know if this is a surprise or not, the version you recently tested is actually *older* than what the bug was first reported with.

Changed in subiquity (Ubuntu):
status: New → Incomplete
Revision history for this message
Kinga Tanska (ktanska) wrote :

Hi,

yes, I know that Ubuntu which we tested is older than the one with defect, but we do have customer who uses it and we are testing it.
I've asked for a retest on Ubuntu 20.04.3 and Ubuntu 22.04.1 with updated subiquity, will be back with the results.

Thanks,
Kinga

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.