You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
For a specified data partition or meta partition , currently, the replica members must use the same raft heartbeat port and replica port, otherwise replicas will not be able to communicate with each other.
func (mp*metaPartition) startRaft() (errerror) {
heartbeatPort, replicaPort, err=mp.getRaftPort() // all peers of this mp use same portfor_, peer:=rangemp.config.Peers {
rp:= raftstore.PeerAddress{
HeartbeatPort: heartbeatPort,
ReplicaPort: replicaPort,
}
peers=append(peers, rp)
}
pc:=&raftstore.PartitionConfig{
Peers: peers,
}
mp.raftPartition, err=mp.config.RaftStore.CreatePartition(pc)
return
}
Why is this needed?
For a single node that only has one net card , in order to deploy cubeFS, we must fake some secondary IPs to satisfy the replicas port requirement.
Anything else?
None yet
The text was updated successfully, but these errors were encountered:
Contact Details
morphes1995@gmail.com
Is there an existing issue for this?
What would you like to be added?
For a specified data partition or meta partition , currently, the replica members must use the same raft heartbeat port and replica port, otherwise replicas will not be able to communicate with each other.
Why is this needed?
For a single node that only has one net card , in order to deploy cubeFS, we must fake some secondary IPs to satisfy the replicas port requirement.
Anything else?
None yet
The text was updated successfully, but these errors were encountered: