HP Matrix 7.2 KVM Private Cloud Backup and Restore
Table Of Contents
- Abstract
 - Matrix Operating Environment with Matrix KVM Private Cloud Overview
 - Backup and restore strategy for Matrix OE with Matrix KVM Private Cloud
 - Matrix KVM Private Cloud backup and restore
 - Matrix KVM Private Cloud Images repository backup and restore
 - Matrix KVM Private Cloud High Availability (HA) cluster configuration backup and restore
 - Appendix A: KVM Private Cloud restore resynchronization actions
 - Appendix B: Images repository restore resynchronization actions
 - Appendix C: Alerts and Audit messages
 - Appendix D: HA Cluster details
 - Appendix E: Backup and Restore REST API
 - Appendix F: Sample Backup Script
 - Appendix G: Sample Restore Script
 - References
 - For more information
 
Validating the HA configuration 
If the cluster needs to be restored for any of the above failure conditions, it is important to validate the restored HA 
configuration after the restore. The following steps outline the recommended validation operation. 
1.  Verify network bonding by observing the output of the following commands.  (This is only needed if a node 
is restored.) 
  ifconfig 
cat /proc/net/bondings/bond<#> 
Check Appendix D for the expected output. 
2.  Verify that the redundant path to storage is valid by checking the status of the multipath service.  
  service multipathd status 
  multipath -ll 
The output should show multiple paths to shared disks (Quorum disk, Images repository disk and Matrix 
KVM Private Cloud appliance disk). Check Appendix D for the expected output. 
3.  Verify that the cluster is running by issuing the following command: 
  clustat 
Check Appendix D for the expected output. 
4.  Verify the logical volume configuration by issuing the following command: 
  lvscan 
Make sure that the logical volumes vg_isc_root and vg_isc_glance are ACTIVE on the node where 
the KVM Private Cloud is running. 
5.  Run virsh list
 to confirm that the KVM Private Cloud is running. 
6.  Validate a manual failover.  
Fail over the KVM Private Cloud by relocating a cluster service on host1. Confirm that the KVM Private 
Cloud failed over to the other node <host2>. 
  clusvcadm –r service:isc_appliance –m <member name 2> 
Failback the KVM Private Cloud from host2 to host1. 
  clusvcadm –r service:isc_appliance –m <member name 1> 
In each of the above cases, issue clustat and verify that the failover and failback happened correctly. 
7.  Simulate a failure condition to test the failover and failback functionality. 
Manually stop the resource manager on host1, assuming the KVM Private Cloud is running on host1. 
  service rgmanager stop 
Verify that the service has been failed over to host2. It may take a few minutes for the failover to complete. 
  clustat 
Restart the resource manager on host1 to verify that the failback functioned properly. 
  service rgmanager start 
Verify that the KVM Private Cloud is running on host1. 
  clustat 
8.  Optionally, you can validate the recovery method “restart” in the cluster configuration as follows: 
•  Log in to the KVM Private Cloud.  
•  Halt the system: 
    shutdown –h now 
• 
Wait for a while. The KVM Private Cloud will reboot and open automatically. 
9.  Check the KVM Private Cloud logs to verify that the VM started without any errors. 
22 










