When using an HDLM device to create a new logical volume

Dynamic Link Manager (for Linux®) User Guide

Version

9.0.x

Audience

anonymous

Part Number

MK-92DLM113-64

ft:lastEdition

2025-11-06

To create a new logical volume on an HDLM device:

This subsection explains the procedure for using an HDLM device to create a new logical volume. Use the following procedure to configure the environment shown in Figure 1.

Figure. Device configuration when creating a logical volume on an HDLM device

Stop all processes and services that are using the HDLM device.
Create a disk partition on the HDLM device and change the system ID from Linux (83 in hexadecimal representation) to Linux LVM (8e in hexadecimal representation).

The following shows an example of how to create two partitions in /dev/sddlmad (HDLM device):

Example of executing the fdisk command (IA32):

Example of executing the parted command (IA32, or EM64T/AMD64):

When you use the parted command to change the partition, to let the system recognize the partitions on the HDLM device, execute the blockdev command after the parted command.

Edit the /etc/lvm/lvm.conf file.

Edit the device section in the file as shown below to disable the SCSI device configuration and enable the HDLM device configuration:

The following shows an example of editing the /etc/lvm/lvm.conf file.

For Red Hat Enterprise Linux 9 or Oracle Linux 9:

Before:

devices {
                              :
        # global_filter = [ "a/.*/" ]
        global_filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
                              :
        # types = [ "fd", 16 ]
                              :
        md_component_detection = 1
                              :
        allow_changes_with_duplicate_pvs = 0
                              :
        multipath_component_detection=1
                              :
        use_devicesfile=1
                              :

}

After:

devices {
                              :
        # global_filter = [ "a/.*/" ]
        # global_filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
        global_filter = [ "a|sddlm[a-p][a-p].*|", "r|/dev/sd|" ]
                              :
        # types = [ "fd", 16 ]
        types = [ "sddlmfdrv", 16 ]
                              :
        # md_component_detection = 1
        md_component_detection=0
                              :
        # allow_changes_with_duplicate_pvs = 0
        allow_changes_with_duplicate_pvs = 1
                              :
        # multipath_component_detection=1
        multipath_component_detection=0
                              :
        # use_devicesfile=1
        use_devicesfile=0
                              :

}

Comment out the existing global_filter, types, md_component_detection, allow_changes_with_duplicate_pvs, multipath_component_detection, and use_devicesfile lines, and then add the underlined parts.

Record the original values, because they are necessary when restoring the SCSI device configuration.

For Red Hat Enterprise Linux 8, Oracle Linux 8, Oracle Unbreakable Enterprise Kernel 8 or SUSE LINUX Enterprise Server 15 SP2 or later:

For Red Hat Enterprise Linux 8.8 or later, Oracle Linux 8.8 or later, or SUSE LINUX Enterprise Server 15 SP5 or later, set the value of multipath_component_detection to 0. For details, see NOTES_ON_USING_LVM2=GUID-91F95B7F-B4F1-4A73-A110-536AD5C219CC=5=EN=.html#GUID-91F95B7F-B4F1-4A73-A110-536AD5C219CC .

Before:

devices {
                              :
        # global_filter = [ "a/.*/" ]
        global_filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
                              :
        # types = [ "fd", 16 ]
                              :
        md_component_detection = 1
                              :
        allow_changes_with_duplicate_pvs = 0
                              :
}

After:

devices {
                              :
        # global_filter = [ "a/.*/" ]
        # global_filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
        global_filter = [ "a|sddlm[a-p][a-p].*|", "r|/dev/sd|" ]
                              :
        # types = [ "fd", 16 ]
        types = [ "sddlmfdrv", 16 ]
                              :
        # md_component_detection = 1
        md_component_detection=0
                              :
        # allow_changes_with_duplicate_pvs = 0
        allow_changes_with_duplicate_pvs = 1
                              :
}

Comment out the existing global_filter, types, md_component_detection and allow_changes_with_duplicate_pvs lines, and then add the underlined parts.

Record the original values, because they are necessary when restoring the SCSI device configuration.

For SUSE LINUX Enterprise Server 12, SUSE LINUX Enterprise Server 15 SP1 or earlier, Red Hat Enterprise Linux 7, Oracle Linux 7 or Oracle Unbreakable Enterprise Kernel 7:

Specify 1 for allow_changes_with_duplicate_pvs and 0 for write_cache_state. For details, see NOTES_ON_USING_LVM2=GUID-91F95B7F-B4F1-4A73-A110-536AD5C219CC=5=EN=.html.

If you want to use use_lvmetad=0 for operation, specify filter, not global_filter.

Before:

# This section allows you to configure which block devices should
# be used by the LVM system.
devices {
                              :

        # filter = [ "a/.*/" ]
          filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
                              :

        write_cache_state = 1
                              :

        # types = [ "fd", 16 ]
                              :

        md_component_detection = 1
                              :
}

After:

# This section allows you to configure which block devices should
# be used by the LVM system.
devices {    
                             :

        # filter = [ "a/.*/" ]
        # filter = [ "r|sddlm[a-p][a-p].*|", "a|/dev/sd|" ]
            filter = [ "a|sddlm[a-p][a-p].*|", "r|/dev/sd|" ]
                              :

        # write_cache_state = 1
        write_cache_state = 0
                              :

        # types = [ "fd", 16 ]
        types = [ "sddlmfdrv", 16 ]
                              :

        # md_component_detection = 1
            md_component_detection=0
                              :
}

Comment out the existing filter, write_cache_state, types and md_component_detection lines, and then add the underlined parts.

Record the original values, because they are necessary when restoring the SCSI device configuration.

Execute the VG scan.

After editing the file, execute the following command.
```
# /sbin/vgscan
```
Delete the LVM cache files.

If the following files exist, delete them:
- /etc/lvm/.cache
- /etc/lvm/cache/.cache

Create the physical volumes.

The following example shows how to define /dev/sddlmad1 and /dev/sddlmad2 as physical volumes:

# pvcreate /dev/sddlmad1
  Physical volume "/dev/sddlmad1" successfully created
# pvcreate /dev/sddlmad2
  Physical volume "/dev/sddlmad2" successfully created

Create a volume group.

In the following example, the command creates the vg01 volume group by using the physical volumes /dev/sddlmad1 and /dev/sddlmad2:
```
# vgcreate vg01 /dev/sddlmad1 /dev/sddlmad2 
  Volume group "vg01" successfully created
```
Create the logical volume.

In the following example, the command creates the logical volume (lvol1: 100 MB) by using the vg01 volume group:
```
# lvcreate -L 100M -n lvol1 vg01
  Logical volume "lvol1" created
```

Create a file system.

The following example shows how to use the mke2fs command to create a file system on the lvol1 logical volume:

# mke2fs /dev/vg01/lvol1
Filesystem label=
OS type: Linux
Block size=1024 (log=0)
Fragment size=1024 (log=0)
25688 inodes, 102400 blocks
5120 blocks (5.00%) reserved for the super user
First data block=1
13 block groups
8192 blocks per group, 8192 fragments per group
1976 inodes per group
Superblock backups stored on blocks:
        8193, 24577, 40961, 57345, 73729

Writing inode tables: done
Writing superblocks and filesystem accounting information: done

This filesystem will be automatically checked every 38 mounts or 180 days, whichever comes first.  Use tune2fs -c or -i to override.

Create the directory to which the logical volume is to be mounted.

The following example shows how to create the /mnt/lvol1 directory:
```
# mkdir /mnt/lvol1
```
Mount the logical volume.

The following example shows how to mount the logical volume to the /mnt/lvol1 directory:
```
# mount /dev/vg01/lvol1 /mnt/lvol1
```