Fail migration from CentOS 8 to Rocky on baremetal

[zykov-as]

I try migrate from CentOS 8.5.2111 to Rocky 8. CentOS installed on bare metal server and disks added from storage with multipath. When I try run migration script, I get error

./migrate2rocky.sh: line 562: dm-2/partition: No such file or directory

Unable to gather EFI data: Can't find disk name or partition number for dm-2.

An error occurred while we were attempting to convert your system to Rocky Linux. Your system may be unstable. Script will now exit to prevent possible damage.

my disks:

# ls -ahlp /dev/disk/by-id | grep "dm-2"
lrwxrwxrwx. 1 root root  10 Jan 12  2022 dm-name-mpatha -> ../../dm-2
lrwxrwxrwx. 1 root root  10 Jan 12  2022 dm-uuid-mpath-360060e80223a330050413a3300000600 -> ../../dm-2
lrwxrwxrwx. 1 root root  10 Jan 12  2022 scsi-360060e80223a330050413a3300000600 -> ../../dm-2
lrwxrwxrwx. 1 root root  10 Jan 12  2022 wwn-0x60060e80223a330050413a3300000600 -> ../../dm-2

dmsetup info /dev/dm-2 
Name: mpatha 
State: ACTIVE 
Read Ahead: 8192 
Tables present: LIVE 
Open count: 3 
Event number: 0 
Major, minor: 253, 2 
Number of targets: 1 
UUID: mpath-360060e80223a330050413a3300000600

# multipath -l
Jul 13 15:59:36 | /etc/multipath.conf line 3, duplicate keyword: find_multipaths
mpathb (360060e80223a330050413a3300000606) dm-0 HITACHI,OPEN-V
size=200G features='0' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=0 status=active
  |- 2:0:1:1  sde 8:64  active undef running
  |- 2:0:0:1  sdc 8:32  active undef running
  |- 12:0:0:1 sdg 8:96  active undef running
  `- 12:0:1:1 sdi 8:128 active undef running
mpatha (360060e80223a330050413a3300000600) dm-2 HITACHI,OPEN-V
size=40G features='0' hwhandler='0' wp=rw
`-+- policy='service-time 0' prio=0 status=active
  |- 2:0:1:0  sdd 8:48  active undef running
  |- 12:0:0:0 sdf 8:80  active undef running
  |- 2:0:0:0  sdb 8:16  active undef running
  `- 12:0:1:0 sdh 8:112 active undef running

What can I do for run this script?

[lilocruz]

It seems that the script is trying to access a partition on dm-2, which it cannot find. Probably the script expects a certain disk layout, or it might not be fully compatible with multipath devices. But I would suggest you check your multipath.conf for any misconfigurations or any issues, as the output you've provided indicates a duplicate keyword warning. You can also verify the disk and partition layout to make sure the EFI partition is present and correctly identified by the system. You can use lsblk or blkid for this.

[m-kaminski]

Hello there'

I'm under impression I have very similar issue

# bash migrate2rocky.sh  -r
migrate2rocky - Begin logging at Wed 19 Jun 2024 11:51:40 AM CDT.

Updating Subscription Management repositories.
Unable to read consumer identity

This system is not registered with an entitlement server. You can use subscription-manager to register.

Removing dnf cache

migrate2rocky.sh: line 562: dm-0/partition: No such file or directory

Unable to gather EFI data: Can't find disk name or partition number for dm-0.

An error occurred while we were attempting to convert your system to Rocky Linux. Your system may be unstable. Script will now exit to prevent possible damage.

 A log of this installation can be found at /var/log/migrate2rocky.log
[root@localhost ~]# cat /etc/redhat-release 
CentOS Stream release 8
[root@localhost ~]# ls -ahlp /dev/disk/by-id | grep "dm-0"
lrwxrwxrwx. 1 root root   10 Jun 19 11:41 dm-name-mpatha -> ../../dm-0
lrwxrwxrwx. 1 root root   10 Jun 19 11:41 dm-uuid-mpath-eui.ffffffffffffffff -> ../../dm-0
lrwxrwxrwx. 1 root root   10 Jun 19 11:41 scsi-eui.ffffffffffffffff -> ../../dm-0
lrwxrwxrwx. 1 root root   10 Jun 19 11:41 wwn-0xui.ffffffffffffffff -> ../../dm-0
[root@localhost ~]# multipath -l
mpatha (eui.ffffffffffffffff) dm-0 NVME,SHGP31-1000GM-2                         
size=932G features='0' hwhandler='0' wp=rw
|-+- policy='service-time 0' prio=0 status=active
| `- 0:1:1:1 nvme0n1 259:2 active undef running
`-+- policy='service-time 0' prio=0 status=enabled
  `- 2:1:1:1 nvme2n1 259:3 active undef running

Is there a known solution or workaround for this issue?