Based on this blog post https://blogs.technet.microsoft.com/uspartner_ts2team/2015/08/26/azure-vm-drive-attachment-limits/ there is a limit on the disk attachment following the model of number of cpus x2. Is there a technical reason why this limit is in place? If you use kubernetes you may not be able to schedule a pod. The scheduler is not aware of this limit.
This was proposed as a workaround https://github.com/khenidak/dysk but I'm wondering why this very low limit exists in the first place.
The number of data disks are directly tied to the size of the VM. For example, if you go here https://docs.microsoft.com/en-us/azure/virtual-machines/windows/sizes you will see that each VM increasing in resources can handle more data disks.
This restraint is mainly built around performance. If you had a virtual machine with only 2 CPU cores and say 10 data disks you would likely run into performance issues as the CPU power and RAM needed to reach out to all those data disks at once could cause your VM to tap out.
The simple solution would be to use larger VM sizes if you need more disks. Or depending on how much space you have Azure can support up to 4TB data disks.
