Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

[ISSUE #144] node abnormal status detection and recovery #145

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

[ISSUE #144] node abnormal status detection and recovery #145

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
43 changes: 37 additions & 6 deletions src/main/java/io/openmessaging/storage/dledger/DLedgerLeaderElector.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -29,6 +29,7 @@
import java.util.ArrayList;
import java.util.Arrays;
import java.util.List;
import java.util.Map;
import java.util.Random;
import java.util.concurrent.CompletableFuture;
import java.util.concurrent.CountDownLatch;
Expand Down Expand Up @@ -111,9 +112,7 @@ public CompletableFuture<HeartBeatResponse> handleHeartBeat(HeartBeatRequest req
return CompletableFuture.completedFuture(new HeartBeatResponse().term(memberState.currTerm()).code(DLedgerResponseCode.UNEXPECTED_MEMBER.getCode()));
}

if (request.getTerm() < memberState.currTerm()) {
return CompletableFuture.completedFuture(new HeartBeatResponse().term(memberState.currTerm()).code(DLedgerResponseCode.EXPIRED_TERM.getCode()));
} else if (request.getTerm() == memberState.currTerm()) {
if (request.getTerm() == memberState.currTerm()) {
if (request.getLeaderId().equals(memberState.getLeaderId())) {
lastLeaderHeartBeatTime = System.currentTimeMillis();
return CompletableFuture.completedFuture(new HeartBeatResponse());
Expand All @@ -124,8 +123,15 @@ public CompletableFuture<HeartBeatResponse> handleHeartBeat(HeartBeatRequest req
//hold the lock to get the latest term and leaderId
synchronized (memberState) {
if (request.getTerm() < memberState.currTerm()) {
if (memberState.isCandidate() && request.isNeedCheckMemberState()) {
logger.warn("[CHECK_MEMBER_STATE] [HandleHeartBeat] remoteId={} need check member state", request.getLeaderId());
memberState.recoveryToFollower(request.getTerm(), request.getLeaderId());
return CompletableFuture.completedFuture(new HeartBeatResponse());
}
return CompletableFuture.completedFuture(new HeartBeatResponse().term(memberState.currTerm()).code(DLedgerResponseCode.EXPIRED_TERM.getCode()));
} else if (request.getTerm() == memberState.currTerm()) {
}

if (request.getTerm() == memberState.currTerm()) {
if (memberState.getLeaderId() == null) {
changeRoleToFollower(request.getTerm(), request.getLeaderId());
return CompletableFuture.completedFuture(new HeartBeatResponse());
Expand Down Expand Up @@ -283,10 +289,12 @@ private void sendHeartbeats(long term, String leaderId) throws Exception {
break;
}

if (x.getCode() == DLedgerResponseCode.NETWORK_ERROR.getCode())
if (x.getCode() == DLedgerResponseCode.NETWORK_ERROR.getCode()) {
memberState.getPeersLiveTable().put(id, Boolean.FALSE);
else
} else {
memberState.getPeersLiveTable().put(id, Boolean.TRUE);
memberState.getPeersTermTable().put(id, x.getTerm());
}

if (memberState.isQuorum(succNum.get())
|| memberState.isQuorum(succNum.get() + notReadyNum.get())) {
Expand All @@ -305,6 +313,7 @@ private void sendHeartbeats(long term, String leaderId) throws Exception {
beatLatch.await(heartBeatTimeIntervalMs, TimeUnit.MILLISECONDS);
if (memberState.isQuorum(succNum.get())) {
lastSuccHeartBeatTime = System.currentTimeMillis();
checkPeersTermTable();
} else {
logger.info("[{}] Parse heartbeat responses in cost={} term={} allNum={} succNum={} notReadyNum={} inconsistLeader={} maxTerm={} peerSize={} lastSuccHeartBeatTime={}",
memberState.getSelfId(), DLedgerUtils.elapsed(startHeartbeatTimeMs), term, allNum.get(), succNum.get(), notReadyNum.get(), inconsistLeader.get(), maxTerm.get(), memberState.peerSize(), new Timestamp(lastSuccHeartBeatTime));
Expand All @@ -320,6 +329,28 @@ private void sendHeartbeats(long term, String leaderId) throws Exception {
}
}

private void checkPeersTermTable() throws Exception {
if (memberState.getSelfId().equals(memberState.getLeaderId())) {
long leaderTerm = memberState.getPeersTermTable().getOrDefault(memberState.getLeaderId(), -1L);
for (Map.Entry<String, Long> entryTerm : memberState.getPeersTermTable().entrySet()) {
if (entryTerm.getKey().equals(memberState.getSelfId())) {
continue;
}

if (entryTerm.getValue() > leaderTerm) {
HeartBeatRequest heartBeatRequest = new HeartBeatRequest();
heartBeatRequest.setGroup(memberState.getGroup());
heartBeatRequest.setLocalId(memberState.getSelfId());
heartBeatRequest.setRemoteId(entryTerm.getKey());
heartBeatRequest.setLeaderId(memberState.getLeaderId());
heartBeatRequest.setNeedCheckMemberState(true);
heartBeatRequest.setTerm(leaderTerm);
dLedgerRpcService.heartBeat(heartBeatRequest);
}
}
}
}

private void maintainAsLeader() throws Exception {
if (DLedgerUtils.elapsed(lastSendHeartBeatTime) > heartBeatTimeIntervalMs) {
long term;
Expand Down
13 changes: 13 additions & 0 deletions src/main/java/io/openmessaging/storage/dledger/MemberState.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -52,6 +52,7 @@ public class MemberState {
private long knownMaxTermInGroup = -1;
private Map<String, String> peerMap = new HashMap<>();
private Map<String, Boolean> peersLiveTable = new ConcurrentHashMap<>();
private Map<String, Long> peersTermTable = new HashMap<>();

private volatile String transferee;
private volatile long termToTakeLeadership = -1;
Expand Down Expand Up @@ -132,6 +133,7 @@ public synchronized void changeToLeader(long term) {
this.role = LEADER;
this.leaderId = selfId;
peersLiveTable.clear();
peersTermTable.clear();
}

public synchronized void changeToFollower(long term, String leaderId) {
Expand All @@ -153,6 +155,13 @@ public synchronized void changeToCandidate(long term) {
transferee = null;
}

public synchronized void recoveryToFollower(long term, String leaderId) {
this.role = FOLLOWER;
this.leaderId = leaderId;
this.currTerm = term;
transferee = null;
}

public String getTransferee() {
return transferee;
}
Expand Down Expand Up @@ -226,6 +235,10 @@ public Map<String, Boolean> getPeersLiveTable() {
return peersLiveTable;
}

public Map<String, Long> getPeersTermTable() {
return peersTermTable;
}

//just for test
public void setCurrTermForTest(long term) {
PreConditions.check(term >= currTerm, DLedgerResponseCode.ILLEGAL_MEMBER_STATE);
Expand Down
9 changes: 9 additions & 0 deletions src/main/java/io/openmessaging/storage/dledger/protocol/HeartBeatRequest.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -18,4 +18,13 @@

public class HeartBeatRequest extends RequestOrResponse {

private boolean needCheckMemberState = false;

public boolean isNeedCheckMemberState() {
return needCheckMemberState;
}

public void setNeedCheckMemberState(boolean needCheckMemberState) {
this.needCheckMemberState = needCheckMemberState;
}
}