replicaSet == vps

[matteo-rama]

it's more a question rather than an issue but maybe i can get some insight on why there are a specific assumption in the cart

in this snippet

{{- if eq (int .Values.replicaCount) 4 }}
  {{- range $i := until (int .Values.replicaCount) }}    # genera 4 mount                                                                                                                                        
  - name: data-rustfs-{{ $i }}                                                                                                                                                                                 
    mountPath: /data/rustfs{{ $i }}           
  {{- end }}                                                                                                                                                                                                     
  {{- else if eq (int .Values.replicaCount) 16 }}
  - name: data                                                                                                                                                                                                   
    mountPath: /data                                                                                                                                                                                           
  {{- end }}   
 

the number of volumes that are given to a specific pod depend on the replica count.
so if replica = 4 i got 4 each with 5 volumes (1 log and 4 data)
if i set 16 i got 16 pod each with 16 volumes.

now, on prem, the disk abstraction layer in my case it's handled via longhorn, i have a dedicated set of worker where longhorn system work, that have storage. longhorn already replicate the data for me, there is replicaset = 3 for every pvc that get requested.

in this scenario, it's still correct to have 4 PVC for eveyr pod? it's gonna to be a huge waste of space i think
What is the best way to handle this scenario?

replicaCount=4, vps=1 (4 drive totali)                                                                                                                                                             
                                                                                                                                                                                                                 
  Pod rustfs-0 ─ PVC0 ─ 1 drive (nodo A)                                                                                                                                                                         
  Pod rustfs-1 ─ PVC1 ─ 1 drive (nodo B)                                                                                                                                                                         
  Pod rustfs-2 ─ PVC2 ─ 1 drive (nodo C)      
  Pod rustfs-3 ─ PVC3 ─ 1 drive (nodo D)   

my idea is to have a scenario that look like that, with every pod with 1 volume and EC:1 does this make sense?

or i should make a "replica:1" of longhornstorage and use it as not replicated data?