Hello,
I have a consul cluster with 3 servers configured. After partial outage, one of the servers fails to start and throws the following error: Failed to start Consul server: Failed to start Raft: failed to load any existing snapshots.
I want this server to just rejoin the rest of the servers.
Q: How many other servers do you have?
Assuming the other servers are running with valid raft data, can you “wipe out” the Consul’s data_dir and then restart the Consul server?
this is just an idea, I’ll let someone else with more “sure” info pitch in! 
I’ve purged data dir and server started indeed but appeared in raft as “non voter”. I restarted it again and got same error as on the beginning. 
Maybe this could help: https://learn.hashicorp.com/consul/day-2-operations/outage#failure-of-a-server-in-a-multi-server-cluster
Is the failed server still part of the cluster? Maybe a force-leave, wipe-out data directory and rejoin will help out.
Thanks for your advice but it turned out this failing server had version 1.2.x installed by mistake, against 1.6.0 version of the rest. Upgrade fixed the problem.