How to check if a VMware datastore is presented on SAN to a Linux NetBackup host or a NetBackup Appliance

Article: 100039119
Last Published: 2020-02-28
Ratings: 7 1
Product(s): Appliances, NetBackup & Alta Data Protection

Description

For VMware backups to use the SAN transport mode, the Datastore LUNs must be accessible to the VM backup host.

Here is how to check if a datastore has been presented to a NetBackup Appliance or a Linux VM backup host :

  1. Find a VMware ESXi host connected to the datastore in question.
  2. Enable SSH on the ESXi host and log into the shell prompt under the root account. For more information, refer to this VMware KB article: https://kb.vmware.com/kb/2004746
  3. Run the following command at the ESXi shell prompt, replacing <Datastore name> with the name of the Datastore in question: #vmkfstools -P -v10 /vmfs/volumes/<Datastore name> | grep device
  4. Observe the output: Logical device: 58cdf60d-ba933b51-4ae6-00505605cc16
  5. Note the Logical device number – this is the same as the Datastore Logical Volume Identifier - LVID.
  6. On the NetBackup Appliance elevated prompt or superuser (root) prompt on a Linux backup host, run the following command filtering the output to find the Logical device value from step 4: 
    #lsblk -l -o +UUID | grep 58cdf60d-ba933b51-4ae6-00505605cc16
  7. If the lsblk command is not available, use the blkid command instead: 
    #blkid | grep 58cdf60d-ba933b51-4ae6-00505605cc16
  8. Observe the output: 
    hostname:/home/maintenance # lsblk -l -o +UUID | grep 58cdf60d-ba933b51-4ae6-00505605cc16
    sdb1 8:17 0 40G 0 part 58cdf60d-ba933b51-4ae6-00505605cc16
  9. If the output is not empty then the Datastore is presented to the backup host over the path(s) shown - /dev/sdb1 in the example above.
  10. If the output is empty then the Datastore is not presented and backups with SAN transport will fail with error status 23.   In case of a backup failure, the same LVID can be seen in VxMS provider log – search for “No path to device”:
    8:21:41.0806 : g_vixInterfaceLogger:libvix.cpp:1800 <DEBUG> : [VFM_ESINFO] 2017-05-05T08:21:41.806-04:00 trivia -[03732] [Originator@6876 sub=Default] Attempting to open LVID:58cdf60e-ed802bbf-66d2-00505605cc16/58cdf60d-ba933b51-4ae6-00505605cc16/1. 08:21:41.0806 : g_vixInterfaceLogger:libvix.cpp:1800 <DEBUG> : [VFM_ESINFO] 2017-05-05T08:21:41.806-04:00 error -[03732] [Originator@6876 sub=Default] No path to device LVID:58cdf60e-ed802bbf-66d2-00505605cc16/58cdf60d-ba933b51-4ae6-00505605cc16/1 found.
  11. In case of VMFS6 Datastore, we can verify LUNs using disk WWID.
  12. In the ESXi host, search the Datastore for naa (Disk WWID) - 
  13. On the ESX search the datastore for the naa (Disk WWID)
    # vmkfstools -P -v10 /vmfs/volumes/<datastore name>/ | grep naa
      <we will get a naa output, like naa.abcxyz123abc>
  14. Next, on the Appliance search for that WWID:
    # lsscsi --scsi_id | grep abcxyz123abc
    <we will see the corresponding output here>
    Example - 
    #lsscsi --scsi_id | grep abcxyz123abc
    [5:0:7:3]    disk    3PARdata VV               3313  /dev/sdazd  369002acdbxyfr0000000000800025409

Was this content helpful?