ck0I finally resurected one of my Proxmox, making my cluster whole again !! 
This node failed 10 monthes ago, on a 2-nodes cluster (+ a quorum device). During this time my services worked fine, but without redundancy. 
I've been able to got almost all my services up again, but I've an Elasticsearch issue.
I've a cluster of 2 elasticsearch + a small master node which can be migrated between hypervisors in case of failure. My data are duplicated between both data nodes, so I had no dataloss. On the Proxmox node which failed, the issue was caused by the drive owning the Elasticsearch data of this node, meaning that now, this node has no data. That's not an issue because of the duplication, but I can't get my node join again the cluster. I guess I should make a propre node removal and add it again ? But when I attempted this, I still got the same error 
Any feddi-techi have an idea ?