Azure load balancer has FIPs that are not zonally replicated by BlueXP

1. Record the existing load balancer configuration prior to making any changes. This can be done via screenshots or with the CLI commands below:
   • FIP’s: az network lb frontend-ip list -g <Rg_Name> --lb-name <LB_Name> -o table
   • Backend pools: az network nic ip-config list --nic-name <NIC_NAME> -g <Rg_Name> -o table
   • Probe ports: az network lb probe list --lb-name <LB_Name> -g <Rg_Grp_Name> -o table
   • LB rules: az network lb rule list --lb-name <LB_Name> -g <Rg_Name> -o table
2. Power off the CVO via the BlueXP interface.
3. Open a Cloud Shell via the Azure portal or an external host with the Azure CLI installed and connected to the subscription  that will be worked in.
4. Delete each LB rule:

az network lb rule delete -g <Rg_Name> --lb-name <Lb_Name> -n clusterMgmtRule
az network lb rule delete -g <Rg_Name> --lb-name <Lb_Name> -dataARule
az network lb rule delete -g <Rg_Name> --lb-name <Lb_Name> -dataBRule

   5. Delete each FIP:

az network lb frontend-ip delete --lb-name <Lb_Name> -g <Rg_Name> -n clusterMgmtFIP
az network lb frontend-ip delete --lb-name <Lb_Name> -g <Rg_Name> -n dataARule
az network lb frontend-ip delete --lb-name <Lb_Name> -g <Rg_Name> -n dataBRule

   6. Create the clusterMgmtFIP:

az network lb frontend-ip create --name svmFIP --resource-group <Rg_Name> --lb-name <Lb_Name> --subnet  "/subscriptions/c560a042-4311-40cf-beb5../subnets/CVO.sn" --private-ip-address x.x.x.x --private-ip-address-  version IPv4 --zone ["1","2","3"]

   7. Create the dataAFIP: 

az network lb frontend-ip create --name dataAFIP --resource-group <Rg_Name> --lb-name <Lb_Name> --subnet  "/subscriptions/c560a042-4311-40cf-beb5../subnets/CVO.sn" --private-ip-address x.x.x.x --private-ip-address-        version IPv4 --zone ["1","2","3"]

   8. Create the dataBFIP:

az network lb frontend-ip create --name dataBFIP --resource-group <Rg_Name> --lb-name <Lb_Name> --subnet "/subscriptions/c560a042-4311-40cf-beb5../subnets/CVO.sn" --private-ip-address x.x.x.x --private-ip-address-version IPv4 --zone ["1","2","3"

   9. Delete the svmRule:

az network lb rule delete -g <Rg_Name> --lb-name <Lb_Name> -n svmRule

   10. Delete the svmFIP:

az network lb frontend-ip delete --lb-name <Lb_Name> -g <Rg_Name> -n svmFIP

   11. Create the svmFIP:

az network lb frontend-ip create --name svmFIP --resource-group <Rg_Name> --lb-name <Lb_Name> --subnet "/subscriptions/c560a042-4311-40cf-beb5../subnets/CVO.sn" --private-ip-address x.x.x.x --private-ip-address-version IPv4 --zone ["1","2","3"]

   12. Create the clusterMgmtRule:

az network lb rule create -g <Rg_Name> --lb-name <Lb_Name> -n clusterMgmtRule --protocol All --frontend-ip-name clusterMgmtFIP --backend-pool-name backendPool --floating-ip true --probe-name clusterMgmtProbe --idle-timeout 30 --frontend-port 0 --backend-port 0

   13. Create the dataARule:

az network lb rule create -g <Rg_Name> --lb-name <Lb_Name> -n dataARule --protocol All --frontend-ip-name dataAFIP --backend-pool-name backendPool --floating-ip true --probe-name dataAProbe --idle-timeout 30 --frontend-port 0 --backend-port 0

   14. Create the dataBRule:

az network lb rule create -g <Rg_Name> --lb-name <Lb_Name> -n dataBRule --protocol All --frontend-ip-name dataBFIP --backend-pool-name backendPool --floating-ip true --probe-name dataBProbe --idle-timeout 30 --frontend-port 0 --backend-port 0

   15.  Create svmRule:

az network lb rule create -g <Rg_Name> --lb-name <Lb_Name> -n svmRule --protocol All --frontend-ip-name svmFIP --backend-pool-name backendPool --floating-ip true --probe-name svmProbe --idle-timeout 30 --frontend-port 0 --backend-port 0

   16. Review the Azure portal load balancer for the CVO where you performed the work and verify the FIP entries you created are reporting as "Zone redundant" in the                                    details of the FIP configuration.

   17.  Verify that each required component for the load balancer configuration matches what it was prior to the changes.

• Note: IP addresses should match those recorded before you started the remediation.​​​​​

 

   18.  Power on the CVO and verify the HA status is normal and all LIFs are home.

Cluster::> cluster show
Cluster::> storage failover show
Cluster::> net int show -is-home false

   19.  If required, send any LIFs that are not home back to the home port.

Cluster::> net int revert *

   20.  Test connectivity  to verify data access to the nodes.

If this issue is exhibited please reach out to NetApp Technical Support to have a case created so this can be tracked.

Azure HA ZRS Frontend IP Zone Configuration Confluence Page