Basic replication implementation.
This commit is contained in:
parent
dfaf67f9ef
commit
4e129930ea
|
@ -22,31 +22,38 @@ proc raftNodeQuorumMin[SmCommandType, SmStateType](node: RaftNode[SmCommandType,
|
|||
if cnt >= (node.peers.len div 2 + node.peers.len mod 2):
|
||||
result = true
|
||||
|
||||
proc raftNodeCheckCommitIndex*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], msg: RaftMessage[SmCommandType, SmStateType]) =
|
||||
withRLock(node.raftStateMutex):
|
||||
if msg.commitIndex > node.commitIndex:
|
||||
let newcommitIndex = min(msg.commitIndex, raftNodeLogIndexGet(node))
|
||||
|
||||
while node.commitIndex < newcommitIndex:
|
||||
node.commitIndex.inc
|
||||
raftNodeApplyLogEntry(node, raftNodeLogEntryGet(node, node.commitIndex))
|
||||
|
||||
proc raftNodeHandleHeartBeat*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], msg: RaftMessage[SmCommandType, SmStateType]):
|
||||
RaftMessageResponse[SmCommandType, SmStateType] =
|
||||
debug "Received heart-beat", node_id=node.id, sender_id=msg.sender_id, node_current_term=node.currentTerm, sender_term=msg.senderTerm
|
||||
result = RaftMessageResponse[SmCommandType, SmStateType](op: rmoAppendLogEntry, senderId: node.id, receiverId: msg.senderId, msgId: msg.msgId, success: false)
|
||||
withRLock(node.raftStateMutex):
|
||||
if node.state == rnsStopped:
|
||||
return
|
||||
|
||||
if msg.senderTerm >= node.currentTerm:
|
||||
raftNodeCancelTimers(node)
|
||||
if node.state == rnsCandidate:
|
||||
raftNodeAbortElection(node)
|
||||
|
||||
result.success = true
|
||||
node.currentTerm = msg.senderTerm
|
||||
node.votedFor = DefaultUUID
|
||||
node.currentLeaderId = msg.senderId
|
||||
|
||||
raftNodeCheckCommitIndex(node, msg)
|
||||
|
||||
raftNodeScheduleElectionTimeout(node)
|
||||
|
||||
proc raftNodeHandleRequestVote*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], msg: RaftMessage[SmCommandType, SmStateType]):
|
||||
RaftMessageResponse[SmCommandType, SmStateType] =
|
||||
result = RaftMessageResponse[SmCommandType, SmStateType](op: rmoRequestVote, msgId: msg.msgId, senderId: node.id, receiverId: msg.senderId, granted: false)
|
||||
withRLock(node.raftStateMutex):
|
||||
if node.state == rnsStopped:
|
||||
return
|
||||
|
||||
if msg.senderTerm > node.currentTerm and node.votedFor == DefaultUUID:
|
||||
if msg.lastLogTerm > raftNodeLogEntryGet(node, raftNodeLogIndexGet(node)).term or
|
||||
(msg.lastLogTerm == raftNodeLogEntryGet(node, raftNodeLogIndexGet(node)).term and msg.lastLogIndex >= raftNodeLogIndexGet(node)):
|
||||
|
@ -85,7 +92,6 @@ proc raftNodeStartElection*[SmCommandType, SmStateType](node: RaftNode[SmCommand
|
|||
)
|
||||
)
|
||||
|
||||
withRLock(node.raftStateMutex):
|
||||
# Wait for votes or voting timeout
|
||||
let all = allFutures(node.votesFuts)
|
||||
await all or raftTimerCreate(node.votingTimeout, proc()=discard)
|
||||
|
@ -114,9 +120,6 @@ proc raftNodeHandleAppendEntries*[SmCommandType, SmStateType](node: RaftNode[SmC
|
|||
RaftMessageResponse[SmCommandType, SmStateType] =
|
||||
result = RaftMessageResponse[SmCommandType, SmStateType](op: rmoAppendLogEntry, senderId: node.id, receiverId: msg.senderId, msgId: msg.msgId, success: false)
|
||||
withRLock(node.raftStateMutex):
|
||||
if node.state == rnsStopped:
|
||||
return
|
||||
|
||||
if msg.senderTerm >= node.currentTerm:
|
||||
raftNodeCancelTimers(node)
|
||||
if node.state == rnsCandidate:
|
||||
|
@ -146,33 +149,48 @@ proc raftNodeHandleAppendEntries*[SmCommandType, SmStateType](node: RaftNode[SmC
|
|||
for entry in msg.logEntries.get:
|
||||
raftNodeLogAppend(node, entry)
|
||||
|
||||
if msg.commitIndex > node.commitIndex:
|
||||
node.commitIndex = min(msg.commitIndex, raftNodeLogIndexGet(node))
|
||||
raftNodeApplyLogEntry(node, raftNodeLogEntryGet(node, node.commitIndex))
|
||||
raftNodeCheckCommitIndex(node, msg)
|
||||
|
||||
result.success = true
|
||||
|
||||
|
||||
proc raftNodeReplicateSmCommand*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], cmd: SmCommandType) =
|
||||
proc raftNodeReplicateSmCommand*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], cmd: SmCommandType): Future[bool] {.async.} =
|
||||
mixin RaftLogEntry, raftTimerCreate
|
||||
|
||||
result = false
|
||||
|
||||
withRLock(node.raftStateMutex):
|
||||
var
|
||||
logEntry: RaftLogEntry[SmCommandType](term: node.currentTerm, data: cmd, entryType: etData)
|
||||
|
||||
raftNodeLogAppend(node, logEntry)
|
||||
|
||||
for peer in node.peers:
|
||||
var
|
||||
msg: RaftMessage[SmCommandType, SmStateType] = RaftMessage[SmCommandType, SmStateType](
|
||||
op: rmoAppendLogEntry, msgId: genUUID(), senderId: node.id, receiverId: peer.id,
|
||||
senderTerm: node.currentTerm, prevLogIndex: raftNodeLogIndexGet(node),
|
||||
prevLogTerm: raftNodeLogEntryGet(node, raftNodeLogIndexGet(node)).term,
|
||||
senderTerm: node.currentTerm, prevLogIndex: raftNodeLogIndexGet(node) - 1,
|
||||
prevLogTerm: raftNodeLogEntryGet(node, raftNodeLogIndexGet(node) - 1).term,
|
||||
commitIndex: node.commitIndex, entries: @[logEntry]
|
||||
)
|
||||
|
||||
node.replicateFuts.add(node.msgSendCallback(msg))
|
||||
|
||||
node.commitIndex.inc
|
||||
raftNodeApplyLogEntry(node, raftNodeLogEntryGet(node, node.commitIndex)) # Apply to state machine
|
||||
let allReplicateFuts = allFutures(node.replicateFuts)
|
||||
await allReplicateFuts or raftTimerCreate(node.appendEntriesTimeout, proc()=discard)
|
||||
if not allReplicateFuts.finished:
|
||||
debug "Raft Node Replication timeout", node_id=node.id
|
||||
|
||||
var replicateCnt = 0
|
||||
for fut in node.replicateFuts:
|
||||
if fut.finished and not fut.cancelled:
|
||||
let resp = RaftMessageResponse[SmCommandType, SmStateType](fut.read)
|
||||
if resp.success:
|
||||
replicateCnt.inc
|
||||
info "Raft Node Replication success", node_id=node.id, sender_id=resp.senderId
|
||||
else:
|
||||
info "Raft Node Replication failed", node_id=node.id, sender_id=resp.senderId
|
||||
|
||||
if replicateCnt >= (node.peers.len div 2 + node.peers.len mod 2):
|
||||
node.commitIndex = raftNodeLogIndexGet(node)
|
||||
raftNodeApplyLogEntry(node, raftNodeLogEntryGet(node, node.commitIndex)) # Apply to state machine
|
||||
result = true
|
|
@ -16,7 +16,7 @@ proc raftNodeLogIndexGet*[SmCommandType, SmStateType](node: RaftNode[SmCommandTy
|
|||
|
||||
proc raftNodeLogEntryGet*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], logIndex: RaftLogIndex): RaftNodeLogEntry[SmCommandType] =
|
||||
if logIndex > 0:
|
||||
result = node.log.logData[logIndex]
|
||||
result = node.log.logData[logIndex - 1]
|
||||
|
||||
proc raftNodeLogAppend*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], logEntry: RaftNodeLogEntry[SmCommandType]) =
|
||||
node.log.logData.add(logEntry)
|
||||
|
|
|
@ -46,7 +46,8 @@ type
|
|||
RaftNodeClientResponseError* = enum
|
||||
rncreSuccess = 0,
|
||||
rncreFail = 1,
|
||||
rncreNotLeader = 2
|
||||
rncreNotLeader = 2,
|
||||
rncreStopped = 3
|
||||
|
||||
RaftNodeClientRequest*[SmCommandType] = ref object
|
||||
op*: RaftNodeClientRequestOps
|
||||
|
|
|
@ -39,7 +39,7 @@ proc new*[SmCommandType, SmStateType](T: type RaftNode[SmCommandType, SmStateTyp
|
|||
peers: RaftNodePeers
|
||||
|
||||
for peerId in peersIds:
|
||||
peers.add(RaftNodePeer(id: peerId, nextIndex: 0, matchIndex: 0, hasVoted: false, canVote: true))
|
||||
peers.add(RaftNodePeer(id: peerId, nextIndex: 1, matchIndex: 0, hasVoted: false, canVote: true))
|
||||
|
||||
result = T(
|
||||
id: id, state: rnsFollower, currentTerm: 0, peers: peers, commitIndex: 0, lastApplied: 0,
|
||||
|
@ -95,15 +95,24 @@ proc raftNodeMessageDeliver*[SmCommandType, SmStateType](node: RaftNode[SmComman
|
|||
# Process Raft Node Client Requests
|
||||
proc raftNodeServeClientRequest*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmStateType], req: RaftNodeClientRequest[SmCommandType]):
|
||||
Future[RaftNodeClientResponse[SmStateType]] {.async, gcsafe.} =
|
||||
|
||||
withRLock(node.raftStateMutex):
|
||||
if not raftNodeIsLeader(node):
|
||||
return RaftNodeClientResponse(nodeId: node.id, error: rncreNotLeader, currentLeaderId: node.currentLeaderId)
|
||||
|
||||
case req.op
|
||||
of rncroExecSmCommand:
|
||||
# TODO: implemenmt command handling
|
||||
discard
|
||||
of rncroRequestSmState:
|
||||
if raftNodeIsLeader(node):
|
||||
|
||||
let resFut = await raftNodeReplicateSmCommand(node, req.smCommand)
|
||||
|
||||
if resFut.read:
|
||||
return RaftNodeClientResponse(nodeId: node.id, error: rncreSuccess, state: raftNodeStateGet(node))
|
||||
else:
|
||||
return RaftNodeClientResponse(nodeId: node.id, error: rncreNotLeader, currentLeaderId: node.currentLeaderId)
|
||||
return RaftNodeClientResponse(nodeId: node.id, error: rncreFail, state: raftNodeStateGet(node))
|
||||
|
||||
of rncroRequestSmState:
|
||||
return RaftNodeClientResponse(nodeId: node.id, error: rncreSuccess, state: raftNodeStateGet(node))
|
||||
|
||||
else:
|
||||
raiseAssert "Unknown client request operation."
|
||||
|
||||
|
@ -164,7 +173,6 @@ proc raftNodeStop*[SmCommandType, SmStateType](node: RaftNode[SmCommandType, SmS
|
|||
# Abort election if in election
|
||||
if node.state == rnsCandidate:
|
||||
raftNodeAbortElection(node)s
|
||||
node.state = rnsStopped
|
||||
# Cancel pending timers (if any)
|
||||
raftNodeCancelTimers(node)
|
||||
|
||||
|
|
|
@ -30,8 +30,7 @@ type
|
|||
rnsUnknown = 0,
|
||||
rnsFollower = 1,
|
||||
rnsCandidate = 2
|
||||
rnsLeader = 3,
|
||||
rnsStopped = 4
|
||||
rnsLeader = 3
|
||||
|
||||
RaftNodeId* = UUID # uuid4 uniquely identifying every Raft Node
|
||||
RaftNodeTerm* = int # Raft Node Term Type
|
||||
|
|
Loading…
Reference in New Issue