mirror of https://github.com/apache/nifi.git
NIFI-8196: When node is reconnected to cluster, ensure that it re-registers for election of cluster coordinator / primary node. On startup, if cluster coordinator is already registered and is 'this node' then register silently as coordinator and do not join the cluster until there is no Cluster Coordinator or another node is elected. This allows the zookeeper session timeout to elapse.
Signed-off-by: Bryan Bende <bbende@apache.org>
This commit is contained in:
Mark Payne
Bryan Bende
parent
b77dbd5030
commit
03fd59eb2f
|
|
@ -98,7 +98,9 @@ the user presses Ctrl-C. At that time, it will initiate shutdown of the applicat
|
|||
To run NiFi in the background, instead run `bin/nifi.sh start`. This will initiate the application to
|
||||
begin running. To check the status and see if NiFi is currently running, execute the command `bin/nifi.sh status`. NiFi can be shutdown by executing the command `bin/nifi.sh stop`.
|
||||
|
||||
Issuing `bin/nifi.sh start` executes the `nifi.sh` script that starts NiFi in the background and then exits. If you want `nifi.sh` to wait for NiFi to finish scheduling all components before exiting, use the `--wait-for-init` flag with an optional timeout specified in seconds: `bin/nifi.sh start --wait-for-init 120`. If the timeout is not provided, the default timeout of 360 seconds will be used.
|
||||
Issuing `bin/nifi.sh start` executes the `nifi.sh` script that starts NiFi in the background and then exits. If you want `nifi.sh` to wait for NiFi to finish scheduling all components before
|
||||
exiting, use the `--wait-for-init` flag with an optional timeout specified in seconds: `bin/nifi.sh start --wait-for-init 120`. If the timeout is not provided, the default timeout of 15 minutes will
|
||||
be used.
|
||||
|
||||
If NiFi was installed with Homebrew, run the commands `nifi start` or `nifi stop` from anywhere in your file system to start or stop NiFi.
|
||||
|
||||
|
|
|
|||
|
|
@ -28,7 +28,7 @@ NOTE: This document is provided with no warranty. All steps have been evaluated
|
|||
|
||||
== Installing Apache NiFi
|
||||
|
||||
Apache NiFi is an easy to use, powerful, and reliable system to process and distribute data. It supports powerful and scalable directed graphs of data routing, transformation, and system mediation logic. In addition to NiFi, there is the NiFi Toolkit, a collection of command-line tools which help perform administrative tasks such as interacting with remote services, generating TLS certificates, managing nodes in a cluster, and encrypting sensitive configuration values.
|
||||
Apache NiFi is an easy to use, powerful, and reliable system to process and distribute data. It supports powerful and scalable directed graphs of data routing, transformation, and system mediation logic. In addition to NiFi, there is the NiFi Toolkit, a collection of command-line tools which help perform administrative tasks such as interacting with remote services, generating TLS certificates, managing nodes in a cluster, and encrypting sensitive configuration values.
|
||||
|
||||
|=======================================================================================================================
|
||||
|Description | Instructions on downloading the Apache NiFi application and Toolkit
|
||||
|
|
@ -179,7 +179,9 @@ drwxr-xr-x 116 alopresto staff 3.6K Apr 6 15:50 lib/
|
|||
+
|
||||
image::nifi-home-dir-listing.png["NiFi Home directory contents"]
|
||||
. At this point, NiFi can be started with default configuration values (available at link:http://localhost:8080/nifi[`http://localhost:8080/nifi`^]).
|
||||
* `./bin/nifi.sh start` -- Starts NiFi. (It takes some time for NiFi to finish scheduling all components. Issuing `bin/nifi.sh start` executes the `nifi.sh` script that starts NiFi in the background and then exits. If you want `nifi.sh` to wait for NiFi to finish scheduling all components before exiting, use the `--wait-for-init` flag with an optional timeout specified in seconds: `bin/nifi.sh start --wait-for-init 120`. If the timeout is not provided, the default timeout of 360 seconds will be used.)
|
||||
* `./bin/nifi.sh start` -- Starts NiFi. (It takes some time for NiFi to finish scheduling all components. Issuing `bin/nifi.sh start` executes the `nifi.sh` script that starts NiFi in the
|
||||
background and then exits. If you want `nifi.sh` to wait for NiFi to finish scheduling all components before exiting, use the `--wait-for-init` flag with an optional timeout specified in seconds:
|
||||
`bin/nifi.sh start --wait-for-init 120`. If the timeout is not provided, the default timeout of 15 minutes will be used.)
|
||||
* `tail -f logs/nifi-app.log` -- Tails the application log which will indicate when the application has started
|
||||
+
|
||||
image::nifi-app-log-ui-available.png["NiFi application log listing available URLS"]
|
||||
|
|
@ -1146,9 +1148,9 @@ NOTE: The `nifi.cluster.load.balance.host=` entry must be manually populated her
|
|||
** `<property name="Connect String"></property>` -> `<property name="Connect String">node1.nifi:2181,node2.nifi:2182,node3.nifi:2183</property>`
|
||||
* `cp node1.nifi/conf/state-management.xml node2.nifi/conf/state-management.xml` -- Copies the modified `state-management.xml` file from `node1` to `node2`
|
||||
* `cp node1.nifi/conf/state-management.xml node3.nifi/conf/state-management.xml` -- Copies the modified `state-management.xml` file from `node1` to `node3`
|
||||
. Update the `authorizers.xml` file. This file contains the *Initial Node Identities* and *Initial User Identities*. The *users* consist of all human users _and_ all nodes in the cluster. The `authorizers.xml` file can be identical on each node (assuming no other changes were made), so the modified file will be copied to the other nodes.
|
||||
. Update the `authorizers.xml` file. This file contains the *Initial Node Identities* and *Initial User Identities*. The *users* consist of all human users _and_ all nodes in the cluster. The `authorizers.xml` file can be identical on each node (assuming no other changes were made), so the modified file will be copied to the other nodes.
|
||||
+
|
||||
NOTE: Each `Initial User Identity` must have a **unique** name (`Initial User Identity 1`, `Initial User Identity 2`, etc.).
|
||||
NOTE: Each `Initial User Identity` must have a **unique** name (`Initial User Identity 1`, `Initial User Identity 2`, etc.).
|
||||
|
||||
* `$EDITOR node1.nifi/conf/authorizers.xml` -- Opens the `authorizers.xml` file in a text editor
|
||||
* Update the following lines:
|
||||
|
|
@ -1179,11 +1181,11 @@ image::authorizers-xml-initial-node-identities.png["authorizers.xml with Initial
|
|||
* Update the following lines:
|
||||
** `nifi.cluster.flow.election.max.wait.time=5 mins` -> `nifi.cluster.flow.election.max.wait.time=1 min` -- Changes the flow election wait time to 1 min, speeding up cluster availability
|
||||
** `nifi.cluster.flow.election.max.candidates=` -> `nifi.cluster.flow.election.max.candidates=3` -- Changes the flow election to occur when 3 nodes are present, speeding up cluster availability
|
||||
. Start the NiFi cluster. Once all three nodes have started and joined the cluster, the flow is agreed upon and a cluster coordinator is elected, the UI is available on any of the three nodes.
|
||||
. Start the NiFi cluster. Once all three nodes have started and joined the cluster, the flow is agreed upon and a cluster coordinator is elected, the UI is available on any of the three nodes.
|
||||
* `./node1.nifi/bin/nifi.sh start` -- Starts `node1`
|
||||
* `./node2.nifi/bin/nifi.sh start` -- Starts `node2`
|
||||
* `./node3.nifi/bin/nifi.sh start` -- Starts `node3`
|
||||
. Wait approximately 30 seconds and open the web browser to link:https://node3.nifi:9443/nifi[`https://node3.nifi:9443/nifi`^]. The cluster should be available. _Note that the Initial Admin Identity has no permissions on the root process group. This is an artifact of legacy design decisions where the root process group ID is not known at initial start time._
|
||||
. Wait approximately 30 seconds and open the web browser to link:https://node3.nifi:9443/nifi[`https://node3.nifi:9443/nifi`^]. The cluster should be available. _Note that the Initial Admin Identity has no permissions on the root process group. This is an artifact of legacy design decisions where the root process group ID is not known at initial start time._
|
||||
+
|
||||
--
|
||||
_The running cluster_
|
||||
|
|
@ -1194,14 +1196,14 @@ _The cluster status from the global menu at the top right_
|
|||
|
||||
image::nifi-secure-cluster-status.png["NiFi secure cluster status pane"]
|
||||
--
|
||||
. To update the Initial Admin Identity's permissions for the root process group, stop each node in the cluster and remove the `authorizations.xml` and `users.xml` file from each node. These files will be regenerated when the node starts again and be populated with the correct permissions.
|
||||
. To update the Initial Admin Identity's permissions for the root process group, stop each node in the cluster and remove the `authorizations.xml` and `users.xml` file from each node. These files will be regenerated when the node starts again and be populated with the correct permissions.
|
||||
* `./node1.nifi/bin/nifi.sh stop` -- Stops `node1`
|
||||
* `rm node1.nifi/conf/authorizations.xml node1.nifi/conf/users.xml` -- Removes the `authorizations.xml` and `users.xml` for `node1`
|
||||
* `./node2.nifi/bin/nifi.sh stop` -- Stops `node2`
|
||||
* `rm node2.nifi/conf/authorizations.xml node2.nifi/conf/users.xml` -- Removes the `authorizations.xml` and `users.xml` for `node2`
|
||||
* `./node3.nifi/bin/nifi.sh stop` -- Stops `node3`
|
||||
* `rm node3.nifi/conf/authorizations.xml node3.nifi/conf/users.xml` -- Removes the `authorizations.xml` and `users.xml` for `node3`
|
||||
. Start the nodes again.
|
||||
. Start the nodes again.
|
||||
* `./node1.nifi/bin/nifi.sh start` -- Starts `node1`
|
||||
* `./node2.nifi/bin/nifi.sh start` -- Starts `node2`
|
||||
* `./node3.nifi/bin/nifi.sh start` -- Starts `node3`
|
||||
|
|
|
|||
|
|
@ -17,10 +17,6 @@
|
|||
|
||||
package org.apache.nifi.cluster.protocol;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.net.InetSocketAddress;
|
||||
import java.net.Socket;
|
||||
|
||||
import org.apache.nifi.cluster.protocol.message.ClusterWorkloadRequestMessage;
|
||||
import org.apache.nifi.cluster.protocol.message.ClusterWorkloadResponseMessage;
|
||||
import org.apache.nifi.cluster.protocol.message.ConnectionRequestMessage;
|
||||
|
|
@ -31,8 +27,16 @@ import org.apache.nifi.cluster.protocol.message.ProtocolMessage;
|
|||
import org.apache.nifi.cluster.protocol.message.ProtocolMessage.MessageType;
|
||||
import org.apache.nifi.io.socket.SocketConfiguration;
|
||||
import org.apache.nifi.io.socket.SocketUtils;
|
||||
import org.slf4j.Logger;
|
||||
import org.slf4j.LoggerFactory;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.net.InetSocketAddress;
|
||||
import java.net.Socket;
|
||||
import java.util.Objects;
|
||||
|
||||
public abstract class AbstractNodeProtocolSender implements NodeProtocolSender {
|
||||
private static final Logger logger = LoggerFactory.getLogger(AbstractNodeProtocolSender.class);
|
||||
private final SocketConfiguration socketConfiguration;
|
||||
private final ProtocolContext<ProtocolMessage> protocolContext;
|
||||
|
||||
|
|
@ -42,10 +46,24 @@ public abstract class AbstractNodeProtocolSender implements NodeProtocolSender {
|
|||
}
|
||||
|
||||
@Override
|
||||
public ConnectionResponseMessage requestConnection(final ConnectionRequestMessage msg) throws ProtocolException, UnknownServiceAddressException {
|
||||
public ConnectionResponseMessage requestConnection(final ConnectionRequestMessage msg, final boolean allowConnectToSelf) throws ProtocolException, UnknownServiceAddressException {
|
||||
Socket socket = null;
|
||||
try {
|
||||
socket = createSocket();
|
||||
final InetSocketAddress socketAddress;
|
||||
try {
|
||||
socketAddress = getServiceAddress();
|
||||
} catch (final IOException e) {
|
||||
throw new ProtocolException("Could not determined address of Cluster Coordinator", e);
|
||||
}
|
||||
|
||||
// If node is not allowed to connect to itself, then we need to check the address of the Cluster Coordinator.
|
||||
// If the Cluster Coordinator is currently set to this node, then we will throw an UnknownServiceAddressException
|
||||
if (!allowConnectToSelf) {
|
||||
validateNotConnectingToSelf(msg, socketAddress);
|
||||
}
|
||||
|
||||
logger.info("Cluster Coordinator is located at {}. Will send Cluster Connection Request to this address", socketAddress);
|
||||
socket = createSocket(socketAddress);
|
||||
|
||||
try {
|
||||
// marshal message to output stream
|
||||
|
|
@ -76,6 +94,21 @@ public abstract class AbstractNodeProtocolSender implements NodeProtocolSender {
|
|||
}
|
||||
}
|
||||
|
||||
private void validateNotConnectingToSelf(final ConnectionRequestMessage msg, final InetSocketAddress socketAddress) {
|
||||
final NodeIdentifier localNodeIdentifier = msg.getConnectionRequest().getProposedNodeIdentifier();
|
||||
if (localNodeIdentifier == null) {
|
||||
return;
|
||||
}
|
||||
|
||||
final String localAddress = localNodeIdentifier.getSocketAddress();
|
||||
final int localPort = localNodeIdentifier.getSocketPort();
|
||||
|
||||
if (Objects.equals(localAddress, socketAddress.getHostString()) && localPort == socketAddress.getPort()) {
|
||||
throw new UnknownServiceAddressException("Cluster Coordinator is currently " + socketAddress.getHostString() + ":" + socketAddress.getPort() + ", which is this node, but " +
|
||||
"connecting to self is not allowed at this phase of the lifecycle. This node must wait for a new Cluster Coordinator to be elected before connecting to the cluster.");
|
||||
}
|
||||
}
|
||||
|
||||
@Override
|
||||
public HeartbeatResponseMessage heartbeat(final HeartbeatMessage msg, final String address) throws ProtocolException {
|
||||
final String hostname;
|
||||
|
|
@ -112,11 +145,9 @@ public abstract class AbstractNodeProtocolSender implements NodeProtocolSender {
|
|||
throw new ProtocolException("Expected message type '" + MessageType.CLUSTER_WORKLOAD_RESPONSE + "' but found '" + responseMessage.getType() + "'");
|
||||
}
|
||||
|
||||
private Socket createSocket() {
|
||||
InetSocketAddress socketAddress = null;
|
||||
private Socket createSocket(final InetSocketAddress socketAddress) {
|
||||
try {
|
||||
// create a socket
|
||||
socketAddress = getServiceAddress();
|
||||
return SocketUtils.createSocket(socketAddress, socketConfiguration);
|
||||
} catch (final IOException ioe) {
|
||||
if (socketAddress == null) {
|
||||
|
|
|
|||
|
|
@ -34,12 +34,13 @@ public interface NodeProtocolSender {
|
|||
* Sends a "connection request" message to the cluster manager.
|
||||
*
|
||||
* @param msg a message
|
||||
* @param allowConnectionToSelf whether or not the node should be allowed to connect to the cluster if the Cluster Coordinator is this node
|
||||
* @return the response
|
||||
* @throws UnknownServiceAddressException if the cluster manager's address
|
||||
* is not known
|
||||
* @throws ProtocolException if communication failed
|
||||
*/
|
||||
ConnectionResponseMessage requestConnection(ConnectionRequestMessage msg) throws ProtocolException, UnknownServiceAddressException;
|
||||
ConnectionResponseMessage requestConnection(ConnectionRequestMessage msg, boolean allowConnectionToSelf) throws ProtocolException, UnknownServiceAddressException;
|
||||
|
||||
/**
|
||||
* Sends a heartbeat to the address given
|
||||
|
|
|
|||
|
|
@ -16,9 +16,6 @@
|
|||
*/
|
||||
package org.apache.nifi.cluster.protocol.impl;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.util.Collection;
|
||||
|
||||
import org.apache.nifi.cluster.protocol.NodeProtocolSender;
|
||||
import org.apache.nifi.cluster.protocol.ProtocolException;
|
||||
import org.apache.nifi.cluster.protocol.ProtocolHandler;
|
||||
|
|
@ -32,6 +29,9 @@ import org.apache.nifi.cluster.protocol.message.HeartbeatMessage;
|
|||
import org.apache.nifi.cluster.protocol.message.HeartbeatResponseMessage;
|
||||
import org.apache.nifi.reporting.BulletinRepository;
|
||||
|
||||
import java.io.IOException;
|
||||
import java.util.Collection;
|
||||
|
||||
public class NodeProtocolSenderListener implements NodeProtocolSender, ProtocolListener {
|
||||
private final NodeProtocolSender sender;
|
||||
private final ProtocolListener listener;
|
||||
|
|
@ -85,8 +85,8 @@ public class NodeProtocolSenderListener implements NodeProtocolSender, ProtocolL
|
|||
}
|
||||
|
||||
@Override
|
||||
public ConnectionResponseMessage requestConnection(final ConnectionRequestMessage msg) throws ProtocolException, UnknownServiceAddressException {
|
||||
return sender.requestConnection(msg);
|
||||
public ConnectionResponseMessage requestConnection(final ConnectionRequestMessage msg, boolean allowConnectToSelf) throws ProtocolException, UnknownServiceAddressException {
|
||||
return sender.requestConnection(msg, allowConnectToSelf);
|
||||
}
|
||||
|
||||
@Override
|
||||
|
|
|
|||
|
|
@ -82,6 +82,7 @@ import java.util.HashMap;
|
|||
import java.util.HashSet;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Objects;
|
||||
import java.util.Set;
|
||||
import java.util.UUID;
|
||||
import java.util.concurrent.ConcurrentHashMap;
|
||||
|
|
@ -553,7 +554,7 @@ public class NodeClusterCoordinator implements ClusterCoordinator, ProtocolHandl
|
|||
@Override
|
||||
public void disconnectionRequestedByNode(final NodeIdentifier nodeId, final DisconnectionCode disconnectionCode, final String explanation) {
|
||||
logger.info("{} requested disconnection from cluster due to {}", nodeId, explanation == null ? disconnectionCode : explanation);
|
||||
updateNodeStatus(new NodeConnectionStatus(nodeId, disconnectionCode, explanation));
|
||||
updateNodeStatus(new NodeConnectionStatus(nodeId, disconnectionCode, explanation), false);
|
||||
|
||||
final Severity severity;
|
||||
switch (disconnectionCode) {
|
||||
|
|
@ -839,7 +840,12 @@ public class NodeClusterCoordinator implements ClusterCoordinator, ProtocolHandl
|
|||
// about a node status from a different node, since those may be received out-of-order.
|
||||
final NodeConnectionStatus currentStatus = updateNodeStatus(nodeId, status);
|
||||
final NodeConnectionState currentState = currentStatus == null ? null : currentStatus.getState();
|
||||
logger.info("Status of {} changed from {} to {}", nodeId, currentStatus, status);
|
||||
if (Objects.equals(status, currentStatus)) {
|
||||
logger.debug("Received notification of Node Status Change for {} but the status remained the same: {}", nodeId, status);
|
||||
} else {
|
||||
logger.info("Status of {} changed from {} to {}", nodeId, currentStatus, status);
|
||||
}
|
||||
|
||||
logger.debug("State of cluster nodes is now {}", nodeStatuses);
|
||||
|
||||
latestUpdateId.updateAndGet(curVal -> Math.max(curVal, status.getUpdateIdentifier()));
|
||||
|
|
|
|||
|
|
@ -2169,6 +2169,8 @@ public class FlowController implements ReportingTaskProvider, Authorizable, Node
|
|||
throw new IllegalStateException("Unable to stop heartbeating because heartbeating is not configured.");
|
||||
}
|
||||
|
||||
LOG.info("Will no longer send heartbeats");
|
||||
|
||||
writeLock.lock();
|
||||
try {
|
||||
if (!isHeartbeating()) {
|
||||
|
|
@ -2348,12 +2350,7 @@ public class FlowController implements ReportingTaskProvider, Authorizable, Node
|
|||
// update the bulletin repository
|
||||
if (isChanging) {
|
||||
if (clustered) {
|
||||
registerForPrimaryNode();
|
||||
|
||||
// Participate in Leader Election for Heartbeat Monitor. Start the heartbeat monitor
|
||||
// if/when we become leader and stop it when we lose leader role
|
||||
registerForClusterCoordinator(true);
|
||||
|
||||
onClusterConnect();
|
||||
leaderElectionManager.start();
|
||||
stateManagerProvider.enableClusterProvider();
|
||||
|
||||
|
|
@ -2382,6 +2379,16 @@ public class FlowController implements ReportingTaskProvider, Authorizable, Node
|
|||
}
|
||||
}
|
||||
|
||||
public void onClusterConnect() {
|
||||
registerForPrimaryNode();
|
||||
|
||||
// Participate in Leader Election for Heartbeat Monitor. Start the heartbeat monitor
|
||||
// if/when we become leader and stop it when we lose leader role
|
||||
registerForClusterCoordinator(true);
|
||||
|
||||
resumeHeartbeats();
|
||||
}
|
||||
|
||||
public void onClusterDisconnect() {
|
||||
leaderElectionManager.unregister(ClusterRoles.PRIMARY_NODE);
|
||||
leaderElectionManager.unregister(ClusterRoles.CLUSTER_COORDINATOR);
|
||||
|
|
|
|||
|
|
@ -23,6 +23,7 @@ import org.apache.nifi.authorization.ManagedAuthorizer;
|
|||
import org.apache.nifi.bundle.Bundle;
|
||||
import org.apache.nifi.cluster.ConnectionException;
|
||||
import org.apache.nifi.cluster.coordination.ClusterCoordinator;
|
||||
import org.apache.nifi.cluster.coordination.node.ClusterRoles;
|
||||
import org.apache.nifi.cluster.coordination.node.DisconnectionCode;
|
||||
import org.apache.nifi.cluster.coordination.node.NodeConnectionState;
|
||||
import org.apache.nifi.cluster.coordination.node.NodeConnectionStatus;
|
||||
|
|
@ -652,13 +653,22 @@ public class StandardFlowService implements FlowService, ProtocolHandler {
|
|||
connectionResponse = connect(false, false, createDataFlowFromController());
|
||||
}
|
||||
|
||||
if (connectionResponse == null) {
|
||||
// If we could not communicate with the cluster, just log a warning and return.
|
||||
// If the node is currently in a CONNECTING state, it will continue to heartbeat, and that will continue to
|
||||
// result in attempting to connect to the cluster.
|
||||
logger.warn("Received a Reconnection Request that contained no DataFlow, and was unable to communicate with an active Cluster Coordinator. Cannot connect to cluster at this time.");
|
||||
controller.resumeHeartbeats();
|
||||
return;
|
||||
}
|
||||
|
||||
loadFromConnectionResponse(connectionResponse);
|
||||
|
||||
clusterCoordinator.resetNodeStatuses(connectionResponse.getNodeConnectionStatuses().stream()
|
||||
.collect(Collectors.toMap(NodeConnectionStatus::getNodeIdentifier, status -> status)));
|
||||
// reconnected, this node needs to explicitly write the inherited flow to disk, and resume heartbeats
|
||||
saveFlowChanges();
|
||||
controller.resumeHeartbeats();
|
||||
controller.onClusterConnect();
|
||||
|
||||
logger.info("Node reconnected.");
|
||||
} catch (final Exception ex) {
|
||||
|
|
@ -871,8 +881,25 @@ public class StandardFlowService implements FlowService, ProtocolHandler {
|
|||
try {
|
||||
logger.info("Connecting Node: " + nodeId);
|
||||
|
||||
// Upon NiFi startup, the node will register for the Cluster Coordinator role with the Leader Election Manager.
|
||||
// Sometimes the node will register as an active participant, meaning that it wants to be elected. This happens when the entire cluster starts up,
|
||||
// for example. (This is determined by checking whether or not there already is a Cluster Coordinator registered).
|
||||
// Other times, it registers as a 'silent' member, meaning that it will not be elected.
|
||||
// If the leader election timeout is long (say 30 or 60 seconds), it is possible that this node was the Leader and was then restarted,
|
||||
// and upon restart found that itself was already registered as the Cluster Coordinator. As a result, it registers as a Silent member of the
|
||||
// election, and then connects to itself as the Cluster Coordinator. At this point, since the node has just restarted, it doesn't know about
|
||||
// any of the nodes in the cluster. As a result, it will get the Cluster Topology from itself, and think there are no other nodes in the cluster.
|
||||
// This causes all other nodes to send in their heartbeats, which then results in them being disconnected because they were previously unknown and
|
||||
// as a result asked to reconnect to the cluster.
|
||||
//
|
||||
// To avoid this, we do not allow the node to connect to itself if it's not an active participant. This means that when the entire cluster is started
|
||||
// up, the node can still connect to itself because it will be an active participant. But if it is then restarted, it won't be allowed to connect
|
||||
// to itself. It will instead have to wait until another node is elected Cluster Coordinator.
|
||||
final boolean activeCoordinatorParticipant = controller.getLeaderElectionManager().isActiveParticipant(ClusterRoles.CLUSTER_COORDINATOR);
|
||||
|
||||
// create connection request message
|
||||
final ConnectionRequest request = new ConnectionRequest(nodeId, dataFlow);
|
||||
|
||||
final ConnectionRequestMessage requestMsg = new ConnectionRequestMessage();
|
||||
requestMsg.setConnectionRequest(request);
|
||||
|
||||
|
|
@ -890,7 +917,7 @@ public class StandardFlowService implements FlowService, ProtocolHandler {
|
|||
ConnectionResponse response = null;
|
||||
for (int i = 0; i < maxAttempts || retryIndefinitely; i++) {
|
||||
try {
|
||||
response = senderListener.requestConnection(requestMsg).getConnectionResponse();
|
||||
response = senderListener.requestConnection(requestMsg, activeCoordinatorParticipant).getConnectionResponse();
|
||||
|
||||
if (response.shouldTryLater()) {
|
||||
logger.info("Requested by cluster coordinator to retry connection in " + response.getTryLaterSeconds() + " seconds with explanation: " + response.getRejectionReason());
|
||||
|
|
@ -907,6 +934,7 @@ public class StandardFlowService implements FlowService, ProtocolHandler {
|
|||
response = null;
|
||||
break;
|
||||
} else {
|
||||
logger.info("Received successful response from Cluster Coordinator to Connection Request");
|
||||
// we received a successful connection response from cluster coordinator
|
||||
break;
|
||||
}
|
||||
|
|
|
|||
|
|
@ -96,6 +96,17 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
logger.info("{} started", this);
|
||||
}
|
||||
|
||||
@Override
|
||||
public boolean isActiveParticipant(final String roleName) {
|
||||
final RegisteredRole role = registeredRoles.get(roleName);
|
||||
if (role == null) {
|
||||
return false;
|
||||
}
|
||||
|
||||
final String participantId = role.getParticipantId();
|
||||
return participantId != null;
|
||||
}
|
||||
|
||||
@Override
|
||||
public void register(String roleName, LeaderElectionStateChangeListener listener) {
|
||||
register(roleName, listener, null);
|
||||
|
|
@ -158,16 +169,18 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
|
||||
final LeaderRole leaderRole = leaderRoles.remove(roleName);
|
||||
if (leaderRole == null) {
|
||||
logger.info("Cannot unregister Leader Election Role '{}' becuase that role is not registered", roleName);
|
||||
logger.info("Cannot unregister Leader Election Role '{}' because that role is not registered", roleName);
|
||||
return;
|
||||
}
|
||||
|
||||
final LeaderSelector leaderSelector = leaderRole.getLeaderSelector();
|
||||
if (leaderSelector == null) {
|
||||
logger.info("Cannot unregister Leader Election Role '{}' becuase that role is not registered", roleName);
|
||||
logger.info("Cannot unregister Leader Election Role '{}' because that role is not registered", roleName);
|
||||
return;
|
||||
}
|
||||
|
||||
leaderRole.getElectionListener().disable();
|
||||
|
||||
leaderSelector.close();
|
||||
logger.info("This node is no longer registered to be elected as the Leader for Role '{}'", roleName);
|
||||
}
|
||||
|
|
@ -233,8 +246,15 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
|
||||
@Override
|
||||
public boolean isLeader(final String roleName) {
|
||||
final boolean activeParticipant = isActiveParticipant(roleName);
|
||||
if (!activeParticipant) {
|
||||
logger.debug("Node is not an active participant in election for role {} so cannot be leader", roleName);
|
||||
return false;
|
||||
}
|
||||
|
||||
final LeaderRole role = getLeaderRole(roleName);
|
||||
if (role == null) {
|
||||
logger.debug("Node is an active participant in election for role {} but there is no LeaderRole registered so this node cannot be leader", roleName);
|
||||
return false;
|
||||
}
|
||||
|
||||
|
|
@ -423,6 +443,10 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
return leaderSelector;
|
||||
}
|
||||
|
||||
public ElectionListener getElectionListener() {
|
||||
return electionListener;
|
||||
}
|
||||
|
||||
public boolean isLeader() {
|
||||
return electionListener.isLeader();
|
||||
}
|
||||
|
|
@ -458,6 +482,7 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
private final String participantId;
|
||||
|
||||
private volatile boolean leader;
|
||||
private volatile Thread leaderThread;
|
||||
private long leaderUpdateTimestamp = 0L;
|
||||
private final long MAX_CACHE_MILLIS = TimeUnit.SECONDS.toMillis(5L);
|
||||
|
||||
|
|
@ -467,6 +492,17 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
this.participantId = participantId;
|
||||
}
|
||||
|
||||
public void disable() {
|
||||
logger.info("Election Listener for Role {} disabled", roleName);
|
||||
setLeader(false);
|
||||
|
||||
if (leaderThread == null) {
|
||||
logger.debug("Election Listener for Role {} disabled but there is no leader thread. Will not interrupt any threads.", roleName);
|
||||
} else {
|
||||
leaderThread.interrupt();
|
||||
}
|
||||
}
|
||||
|
||||
public synchronized boolean isLeader() {
|
||||
if (leaderUpdateTimestamp < System.currentTimeMillis() - MAX_CACHE_MILLIS) {
|
||||
try {
|
||||
|
|
@ -483,6 +519,8 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
|
||||
return false;
|
||||
}
|
||||
} else {
|
||||
logger.debug("Checking if this node is leader for role {}: using cached response, returning {}", roleName, leader);
|
||||
}
|
||||
|
||||
return leader;
|
||||
|
|
@ -531,6 +569,7 @@ public class CuratorLeaderElectionManager implements LeaderElectionManager {
|
|||
|
||||
@Override
|
||||
public void takeLeadership(final CuratorFramework client) throws Exception {
|
||||
leaderThread = Thread.currentThread();
|
||||
setLeader(true);
|
||||
logger.info("{} This node has been elected Leader for Role '{}'", this, roleName);
|
||||
|
||||
|
|
|
|||
|
|
@ -49,6 +49,13 @@ public interface LeaderElectionManager {
|
|||
*/
|
||||
void register(String roleName, LeaderElectionStateChangeListener listener, String participantId);
|
||||
|
||||
/**
|
||||
* Indicates whether or not this node is an active participant in the election for the given role
|
||||
* @param roleName the name of the role
|
||||
* @return <code>true</code> if this node is an active participant in the election (is allowed to be elected) or <code>false</code> otherwise
|
||||
*/
|
||||
boolean isActiveParticipant(String roleName);
|
||||
|
||||
/**
|
||||
* Returns the Participant ID of the node that is elected the leader, if one was provided when the node registered
|
||||
* for the role via {@link #register(String, LeaderElectionStateChangeListener, String)}. If there is currently no leader
|
||||
|
|
|
|||
|
|
@ -40,6 +40,11 @@ public class StandaloneLeaderElectionManager implements LeaderElectionManager {
|
|||
public void register(final String roleName, final LeaderElectionStateChangeListener listener, final String participantId) {
|
||||
}
|
||||
|
||||
@Override
|
||||
public boolean isActiveParticipant(final String roleName) {
|
||||
return true;
|
||||
}
|
||||
|
||||
@Override
|
||||
public String getLeader(final String roleName) {
|
||||
return null;
|
||||
|
|
|
|||
|
|
@ -236,8 +236,8 @@ public class ZooKeeperStateProvider extends AbstractStateProvider {
|
|||
return zooKeeperClientConfig;
|
||||
} else {
|
||||
Properties stateProviderProperties = new Properties();
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_SESSION_TIMEOUT, String.valueOf(timeoutMillis));
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_CONNECT_TIMEOUT, String.valueOf(timeoutMillis));
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_SESSION_TIMEOUT, timeoutMillis + " millis");
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_CONNECT_TIMEOUT, timeoutMillis + " millis");
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_ROOT_NODE, rootNode);
|
||||
stateProviderProperties.setProperty(NiFiProperties.ZOOKEEPER_CONNECT_STRING, connectionString);
|
||||
zooKeeperClientConfig = ZooKeeperClientConfig.createConfig(combineProperties(nifiProperties, stateProviderProperties));
|
||||
|
|
|
|||
|
|
@ -16,25 +16,6 @@
|
|||
*/
|
||||
package org.apache.nifi.tests.system.clustering;
|
||||
|
||||
import static org.junit.Assert.assertEquals;
|
||||
import static org.junit.Assert.assertFalse;
|
||||
|
||||
import java.io.ByteArrayOutputStream;
|
||||
import java.io.File;
|
||||
import java.io.FileInputStream;
|
||||
import java.io.IOException;
|
||||
import java.io.InputStream;
|
||||
import java.io.StringReader;
|
||||
import java.nio.charset.StandardCharsets;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Properties;
|
||||
import java.util.Set;
|
||||
import java.util.zip.GZIPInputStream;
|
||||
import javax.xml.parsers.DocumentBuilder;
|
||||
import javax.xml.parsers.ParserConfigurationException;
|
||||
import org.apache.nifi.controller.serialization.FlowEncodingVersion;
|
||||
import org.apache.nifi.controller.serialization.FlowFromDOMFactory;
|
||||
import org.apache.nifi.encrypt.StringEncryptor;
|
||||
|
|
@ -69,6 +50,27 @@ import org.w3c.dom.Element;
|
|||
import org.xml.sax.InputSource;
|
||||
import org.xml.sax.SAXException;
|
||||
|
||||
import javax.xml.parsers.DocumentBuilder;
|
||||
import javax.xml.parsers.ParserConfigurationException;
|
||||
import java.io.ByteArrayOutputStream;
|
||||
import java.io.File;
|
||||
import java.io.FileFilter;
|
||||
import java.io.FileInputStream;
|
||||
import java.io.IOException;
|
||||
import java.io.InputStream;
|
||||
import java.io.StringReader;
|
||||
import java.nio.charset.StandardCharsets;
|
||||
import java.util.ArrayList;
|
||||
import java.util.Arrays;
|
||||
import java.util.List;
|
||||
import java.util.Map;
|
||||
import java.util.Properties;
|
||||
import java.util.Set;
|
||||
import java.util.zip.GZIPInputStream;
|
||||
|
||||
import static org.junit.Assert.assertEquals;
|
||||
import static org.junit.Assert.assertFalse;
|
||||
|
||||
public class JoinClusterWithDifferentFlow extends NiFiSystemIT {
|
||||
@Override
|
||||
protected NiFiInstanceFactory getInstanceFactory() {
|
||||
|
|
@ -104,10 +106,18 @@ public class JoinClusterWithDifferentFlow extends NiFiSystemIT {
|
|||
}
|
||||
|
||||
|
||||
private File getBackupFile(final File confDir) {
|
||||
final File[] flowXmlFileArray = confDir.listFiles(file -> file.getName().startsWith("flow") && file.getName().endsWith(".xml.gz"));
|
||||
private File getBackupFile(final File confDir) throws InterruptedException {
|
||||
final FileFilter fileFilter = file -> file.getName().startsWith("flow") && file.getName().endsWith(".xml.gz");
|
||||
|
||||
waitFor(() -> {
|
||||
final File[] flowXmlFileArray = confDir.listFiles(fileFilter);
|
||||
return flowXmlFileArray != null && flowXmlFileArray.length == 2;
|
||||
});
|
||||
|
||||
final File[] flowXmlFileArray = confDir.listFiles(fileFilter);
|
||||
final List<File> flowXmlFiles = new ArrayList<>(Arrays.asList(flowXmlFileArray));
|
||||
assertEquals(2, flowXmlFiles.size());
|
||||
|
||||
flowXmlFiles.removeIf(file -> file.getName().equals("flow.xml.gz"));
|
||||
|
||||
assertEquals(1, flowXmlFiles.size());
|
||||
|
|
|
|||
|
|
@ -50,7 +50,6 @@ import static org.junit.Assert.assertSame;
|
|||
import static org.junit.Assert.assertTrue;
|
||||
|
||||
public class ParameterContextIT extends NiFiSystemIT {
|
||||
|
||||
@Test
|
||||
public void testCreateParameterContext() throws NiFiClientException, IOException {
|
||||
final Set<ParameterEntity> parameterEntities = new HashSet<>();
|
||||
|
|
@ -175,13 +174,17 @@ public class ParameterContextIT extends NiFiSystemIT {
|
|||
setParameterContext("root", createdContextEntity);
|
||||
waitForValidProcessor(processorId);
|
||||
|
||||
// Create a Parameter that sets the 'foo' value to null and denote that the parameter's value should be explicitly removed.
|
||||
final ParameterEntity fooNull = createParameterEntity("foo", null, false, null);
|
||||
fooNull.getParameter().setValueRemoved(true);
|
||||
createdContextEntity.getComponent().setParameters(Collections.singleton(fooNull));
|
||||
getNifiClient().getParamContextClient().updateParamContext(createdContextEntity);
|
||||
|
||||
// Should become invalid because property is required and has no default
|
||||
waitForInvalidProcessor(processorId);
|
||||
}
|
||||
|
||||
|
||||
@Test
|
||||
public void testParametersReferencingEL() throws NiFiClientException, IOException, InterruptedException {
|
||||
final ProcessorEntity generate = getClientUtil().createProcessor("GenerateFlowFile");
|
||||
|
|
|
|||
Loading…
Reference in New Issue