Steps to recreate/rebuild SF cluster by retaining others resources(Eg: Load balancer, Public IP, Vnet and So on…) in the Resource Group when cluster state is lost due to some unsupported operations.
Step 1: Go to Resource Explorer, navigate to SF resource and copy the configuration content of SF resource from the right plane
Step 2: Delete the SF resource in the Resource group:
Step 3: Deploy a Custom Template:
Paste the copied SF configuration in Step 1 under the “resource : [ // Please copy here]
Remove below properties:
- Cluster ID
- etag
- id
- Complete "systemData' info
Add the API version under the type and modify the UpgradeMode to “Manual”
- "apiVersion": "2021-06-01", (Choose latest one from https://learn.microsoft.com/en-us/azure/templates/microsoft.servicefabric/clusters?pivots=deployment-language-arm-template , current version as of 4th April'23 is '2021-06-01')
- "upgradeMode": "Manual"
Save and Create the Resource, Post which SF resource gets created under Resource group.
Step 4: Copy the new Cluster Id from Resource Explorer -> SF resource:
Navigate to VMSS resource -> Replace the Cluster ID in ClusterEndpoint
However VMSS update may result in below failure, Please ignore the Error.
If you go to Resource Group- > VMSS -> ClusterEndpoint will be updated with new Cluster ID
Step 5: Scale the instance count by + 1( In case of Reliability is None, +3 if bronze , +5 if Silver and so on)
Once Instance is created, SF Status would show as “Deploying” and newly added instances will start showing under Nodes.
Step 6: Delete the old instance:
Microsoft