Bug 1411600

$ sudo virsh pool-start vg
error: Failed to start pool vg
error: unsupported configuration: cannot find any matching source devices for logical volume group 'vg'


journalctl -f reports

Jan 09 21:54:52 f25h sudo[14940]:    chris : TTY=pts/2 ; PWD=/home/chris ; USER=root ; COMMAND=/bin/virsh pool-start vg
Jan 09 21:54:52 f25h audit[14940]: USER_CMD pid=14940 uid=1000 auid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='cwd="/home/chris" cmd=766972736820706F6F6C2D7374617274207667 terminal=pts/2 res=success'
Jan 09 21:54:52 f25h audit[14940]: CRED_REFR pid=14940 uid=0 auid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='op=PAM:setcred grantors=pam_env,pam_fprintd acct="root" exe="/usr/bin/sudo" hostname=? addr=? terminal=/dev/pts/2 res=success'
Jan 09 21:54:52 f25h sudo[14940]: pam_systemd(sudo:session): Cannot create session: Already occupied by a session
Jan 09 21:54:52 f25h audit[14940]: USER_START pid=14940 uid=0 auid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='op=PAM:session_open grantors=pam_keyinit,pam_limits,pam_keyinit,pam_limits,pam_systemd,pam_unix acct="root" exe="/usr/bin/sudo" hostname=? addr=? terminal=/dev/pts/2 res=success'
Jan 09 21:54:52 f25h sudo[14940]: pam_unix(sudo:session): session opened for user root by (uid=0)
Jan 09 21:54:52 f25h libvirtd[4511]: unsupported configuration: cannot find any matching source devices for logical volume group 'vg'
Jan 09 21:54:52 f25h sudo[14940]: pam_unix(sudo:session): session closed for user root
Jan 09 21:54:52 f25h audit[14940]: USER_END pid=14940 uid=0 auid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='op=PAM:session_close grantors=pam_keyinit,pam_limits,pam_keyinit,pam_limits,pam_systemd,pam_unix acct="root" exe="/usr/bin/sudo" hostname=? addr=? terminal=/dev/pts/2 res=success'
Jan 09 21:54:52 f25h audit[14940]: CRED_DISP pid=14940 uid=0 auid=1000 ses=3 subj=unconfined_u:unconfined_r:unconfined_t:s0-s0:c0.c1023 msg='op=PAM:setcred grantors=pam_env,pam_fprintd acct="root" exe="/usr/bin/sudo" hostname=? addr=? terminal=/dev/pts/2 res=success'


virsh # pool-dumpxml vg
<pool type='logical'>
  <name>vg</name>
  <uuid>b7598e33-ecb1-4059-baa8-04f658f4a627</uuid>
  <capacity unit='bytes'>0</capacity>
  <allocation unit='bytes'>0</allocation>
  <available unit='bytes'>0</available>
  <source>
    <device path='/dev/nvme0n1p5'/>
    <name>vg</name>
    <format type='lvm2'/>
  </source>
  <target>
    <path>/dev/vg</path>
  </target>
</pool>

Haha, user error!
/dev/nvme0n1p5 is wrong; a recent partitioning change changed this to nvme0n1p7. Seems to me the source device path is superfluous if the pool references an LVM VG UUID which in turn knows what its PVs are. But OK...

I have the same problem and I don't think I am making an error.

- libvirt-2.0.0-10.el7_3.2

The problem occured after adding a new LVM volume to a pool. The storage pools became inactive.

lvcreate -L 96G -n rdvdb-test_disk_E vg_guest_lvmfast

virsh # pool-list --all
 Name                 State      Autostart 
-------------------------------------------
 default              active     yes       
 IsoImages            active     yes       
 SlowDiskOnFile       active     yes       
 SsdDiskOnFile        active     yes       
 vg_guest_lvmfast     inactive   yes       
 vg_guest_lvmslow     inactive   yes       

virsh # pool-start vg_guest_lvmfast                                                                                                    
error: Failed to start pool vg_guest_lvmfast                                                                                           
error: unsupported configuration: cannot find any matching source devices for logical volume group 'vg_guest_lvmfast'                  


virsh # pool-dumpxml vg_guest_lvmfast
<pool type='logical'>                                                                                                                  
  <name>vg_guest_lvmfast</name>                                                                                                        
  <uuid>d0c80fc1-c12e-4ed2-be7c-d8b84f2a48be</uuid>
  <capacity unit='bytes'>0</capacity>
  <allocation unit='bytes'>0</allocation>
  <available unit='bytes'>0</available>
  <source>
    <device path='/dev/vg_guest_lvmfast'/>
    <name>vg_guest_lvmfast</name>
    <format type='lvm2'/>
  </source>
  <target>
    <path>/dev/vg_guest_lvmfast</path>
  </target>
</pool>



vgdisplay vg_guest_lvmfast
  --- Volume group ---
  VG Name               vg_guest_lvmfast
  System ID             
  Format                lvm2
  Metadata Areas        1
  Metadata Sequence No  9
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                4
  Open LV               4
  Max PV                0
  Cur PV                1
  Act PV                1
  VG Size               1,09 TiB
  PE Size               4,00 MiB
  Total PE              285887
  Alloc PE / Size       97024 / 379,00 GiB
  Free  PE / Size       188863 / 737,75 GiB
  VG UUID               KLcujo-eqL9-2D1e-dt74-8aP8-1xar-CdnbFN



ls -l /dev/vg_guest_lvmfast/
total 0
lrwxrwxrwx 1 root root 7 17. Jan 14:18 lv_kvm_aistatistics_diskE -> ../dm-2
lrwxrwxrwx 1 root root 7 17. Jan 14:18 lv_kvm_aistatistics_diskF -> ../dm-3
lrwxrwxrwx 1 root root 7 17. Jan 14:56 rdvdb-test_disk_D -> ../dm-4
lrwxrwxrwx 1 root root 7 17. Jan 14:03 rdvdb-test_disk_E -> ../dm-5


using virsh edit I can add LVM (or any other) disks to VMs, that works fine. 

Best regards,
Arie.

The problem is that libvirt enforces the mapping of the physical volume names to the volume group also after it was built which is not entirely correct. LVM is able to properly detect the presence of all PVs and does not need such care.

As a workaround I'd suggest you delete the physical volume paths from the storage pool definition, then the check will be skipped and only the volume group presence will be checked.

Hi Peter,

thanks for your reply and suggestion.

When you say "delete the physical volume paths from the storage pool definition", then you mean to delete the "device path" XML tag from the "source" container?

best,
Arie.

Yes, exactly. That data is not required besides "building" and deleting of the pool. For regular usage, the name should be sufficient.

Peter,

I commented the "device path" xml tag which caused a complete removal of the line after a libvirtd restart. Well ok.

The result:

virsh # pool-dumpxml vg_guest_lvmfast
<pool type='logical'>
  <name>vg_guest_lvmfast</name>
  <uuid>d0c80fc1-c12e-4ed2-be7c-d8b84f2a48be</uuid>
  <capacity unit='bytes'>1199096987648</capacity>
  <allocation unit='bytes'>406948151296</allocation>
  <available unit='bytes'>792148836352</available>
  <source>
    <name>vg_guest_lvmfast</name>
    <format type='lvm2'/>
  </source>
  <target>
    <path>/dev/vg_guest_lvmfast</path>
  </target>
</pool>

virsh # pool-list --all
 Name                 State      Autostart 
-------------------------------------------
 default              active     yes
 IsoImages            active     yes
 SlowDiskOnFile       active     yes
 SsdDiskOnFile        active     yes
 vg_guest_lvmfast     active     yes
 vg_guest_lvmslow     active     yes


Your workaround worked around!

Thanks again,
Arie.

Thank you for reporting this issue to the libvirt project. Unfortunately we have been unable to resolve this issue due to insufficient maintainer capacity and it will now be closed. This is not a reflection on the possible validity of the issue, merely the lack of resources to investigate and address it, for which we apologise. If you none the less feel the issue is still important, you may choose to report it again at the new project issue tracker https://gitlab.com/libvirt/libvirt/-/issues The project also welcomes contribution from anyone who believes they can provide a solution.