We have a SF cluster with VMSS and a 5-node instance on size StandardA2m_V2, but the disk storage on these vm's is running out of space. We're planning to add a datadisk to these VM's and set the sf-datafolder to these datadisks.
At first I tried to add datadisks to our VMSS. But a few months ago when we setup the cluster, we configured the scale set with unmanaged disks. To add datadisks to the vm's, we had to change the nodes. Updating the resources.azure.com with Azure virtual machine scale sets and attached data disks didn't work out because of the unmanaged disks.
Then I tried to add nodetypes to our existing cluster with PowerShell, and use 'Add-AzureRmServiceFabricNodeType' from this link: How to add new Node Type to deployed Service Fabric cluster?. But then I can't pass my specific VM configuration with datadisks. Nodetype is just a string that is used for naming. And the vmsku parameter is for the 'size'.
I also found Azure Service Fabric node types and virtual machine scale sets but ending up in 'how to' then...
I managed to setup a new cluster, via armtemplate, with the specific configuration in VMSS and every VM with it's own datadisk. But actually I want to update my existing cluster. Any ideas?