Strange errors during backup OpenVZ

[johnplv]

I have a cluster of 3 nodes (Proxmox 3.0) with different VMs on OpenVZ and KVM.
Openvz use local storage, KVM use a iSCSI disks created and connected via NexentaStor API.
Everything works fine, but during backup OpenVZ container in logs appear strange errors related to the ISCSI disks:

INFO: starting new backup job: vzdump 100 101 1010 1110 1030 1000 1020 --quiet 1 --mode snapshot --compress gzip --storage stor2-NFS
INFO: skip external VMs: 1010, 1110, 1030, 1020
INFO: Starting Backup of VM 100 (openvz)
INFO: CTID 100 exist mounted running
INFO: status = running
INFO: backup mode: snapshot
INFO: ionice priority: 7
INFO: creating lvm snapshot of /dev/mapper/pve-data ('/dev/pve/vzsnap-pve-node2-0')
INFO:   /dev/sdb5: read failed after 0 of 4096 at 214489300992: Input/output error
INFO:   /dev/sdb5: read failed after 0 of 4096 at 214489358336: Input/output error
INFO:   Found duplicate PV uUbBSKcLMOKvFzFlA5e7P1ZdcNeaJCXe: using /dev/sdi5 not /dev/sdb5
INFO:   /dev/sdj5: read failed after 0 of 4096 at 34101723136: Input/output error
INFO:   /dev/sdj5: read failed after 0 of 4096 at 34101780480: Input/output error
INFO:   Found duplicate PV iHn0pSJ6CJE3z177RXkoMg6xfTpuQxdF: using /dev/sdn2 not /dev/sdm2
INFO:   Logical volume "vzsnap-pve-node2-0" created
INFO: creating archive '/mnt/pve/stor2-NFS/dump/vzdump-openvz-100-2013_08_07-01_00_02.tar.gz'
INFO: Total bytes written: 57655429120 (54GiB, 13MiB/s)
INFO: archive file size: 33.29GB
INFO: Finished Backup of VM 100 (01:13:05)

What is this? and how to fix it?