Vault: Error initializing storage of type raft: failed to create fsm: timeout

[tsarna]

Describe the bug
I've been playing with the raft storage in 1.3.1. In a 3 node cluster, after restarting the original node (to which the other two were joined), it started logging:

Error initializing storage of type raft: failed to create fsm: timeout

All nodes seem functional however, and in "vault operator raft configuration" doesn't show anything different about this node.

To Reproduce
create 3 nodes master01[abc], with vault 1.3.1. These are running alpine linux with the vault package from the "edge" repo.

Initialize and configure master01a -- various auth mechanism, secrets engines, policies, etc -- I can detail it all if needed but I have no reason to believe the details are relevant.

TLS is disabled.

Join master01b and master01c to master01a.

Some time later, I restarted master01a, then the leader, and it began loging the message. However, it appears to be functional. vault operator raft configuration says:

```Key Value
--- -----
config map[index:14343 servers:[map[address:192.168.16.4:8201 leader:false node_id:master01a protocol_version:3 voter:true] map[address:192.168.18.4:8201 leader:true node_id:master01c protocol_version:3 voter:true] map[address:192.168.17.4:8201 leader:false node_id:master01b protocol_version:3 voter:true]]]

**Expected behavior**
No scary error messages are logged after a restart.

If there are error messages, they should provide more detail to understand what to do to fix them, or should be documented in this regard. I could find nothing useful in searches.

**Environment:**
* Vault Server Version (retrieve with `vault status`): 1.3.1
* Vault CLI Version (retrieve with `vault version`): Vault v1.3.1
* Server Operating System/Architecture: Alpine Linux 3.11.0, x86_64, on AWS t3a.nano instances

Vault server configuration file(s):

```hcl
storage "raft" {
  path = "/data/vault/db"
  node_id = "master01a"
}

listener "tcp" {
  address = "192.168.16.4:8200"
  tls_disable = true
}

listener "tcp" {
  address = "127.0.0.1:8200"
  tls_disable = true
}

seal "awskms" {
  region     = "us-east-1"
  kms_key_id = "xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx"
}

api_addr = "http://192.168.16.4:8200"
cluster_addr = "http://192.168.16.4:8201"
pid_file = "/data/vault/vault.pid"       

01b and 01c are the same but in 192.168.17 and 192.168.18 respectively.

Additional context
Add any other context about the problem here.

[tsarna]

After restarting more recently, I can no longer reproduce this. I'm not sure what changed.

Still, it would be nice if the message and/or documentation added some guidance on what the problem was and anything that could be done to fix it or debug the issue further.