Cisco Cisco Elastic Services Controller 2.3 Guide Du Développeur

Page de 85
    </vm_source> 
    <vm_target> 
      <vmid>22d210a1-15e1-4e2b-a7a9-9796ec7b259c</vmid> 
      <hostid>4f926bb44423804ea8d87cac946b80e637c315f4f606bfa788b19b8c</hostid> 
      <hostname>my-ucs-7</hostname> 
      <interfaces> 
        <interface> 
          <nicid>0</nicid> 
          <type>virtual</type> 
          <port_id>71fd1601-f78b-41ce-a518-48fae917affb</port_id> 
          <network>943fda9e-79f8-400c-b442-3506f102721a</network> 
          <subnet>e313b95c-ca1f-4c81-8d60-c9e721a85d0b</subnet> 
          <ip_address>192.168.0.75</ip_address> 
          <mac_address>fa:16:3e:e9:23:f7</mac_address> 
          <netmask>255.255.255.0</netmask> 
          <gateway>192.168.0.1</gateway> 
        </interface> 
      </interfaces> 
    </vm_target> 
    <event> 
      <type>VM_RECOVERY_COMPLETE</type> 
    </event> 
  </escEvent> 
</notification> 
 
7  ESC Triggered Workflows 
 
7.1  Recovery (Healing) Workflow  
 
A VM recovery is triggered in the following scenarios. 
•  During a service deployment if VM_ALIVE event fromany or all of the VMs being deployed by 
the service is not received before the timeout value. 
•  After the service is deployed, if any of the VMs deployed goes down for some reason and there 
are rules defined to activate VM recovery for these cases. 
Recovery Initialize 
When a VM recovery is triggered, an escEvent of type VM_RECOVERY_INIT with a status of SUCCESS will 
be sent to Netconf subscribers. This indicates that the recovery workflow has been started. This will 
hard reboot the VM instance and checks if the VM comes back up.  
If the VM comes up successfully, an escEvent of type VM_RECOVERY_COMPLETE with a status of 
SUCCESS with be sent. 
Attempts for recovery 
If the VM did not come up after this attempt, then ESC tries to undeploy and redeploy the VM and 
corresponding  escEvent of type VM_RECOVERY_UNDEPLOYED and VM_RECOVERY_DEPLOYED will be 
sent. If the undeploy and redeploy also fails to get the VM back up, this undeploy and redeploy cycle will 
be retried to a maximum value of “
vm_recovery_retries_max
 times which is configurable.