Examples with RMNodeReconnectEvent org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent used on opensource projects

Example 1 with RMNodeReconnectEvent

use of org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent in project hadoop by apache.

the class ResourceTrackerService method registerNodeManager.

@SuppressWarnings("unchecked")
@Override
public RegisterNodeManagerResponse registerNodeManager(RegisterNodeManagerRequest request) throws YarnException, IOException {
    NodeId nodeId = request.getNodeId();
    String host = nodeId.getHost();
    int cmPort = nodeId.getPort();
    int httpPort = request.getHttpPort();
    Resource capability = request.getResource();
    String nodeManagerVersion = request.getNMVersion();
    Resource physicalResource = request.getPhysicalResource();
    RegisterNodeManagerResponse response = recordFactory.newRecordInstance(RegisterNodeManagerResponse.class);
    if (!minimumNodeManagerVersion.equals("NONE")) {
        if (minimumNodeManagerVersion.equals("EqualToRM")) {
            minimumNodeManagerVersion = YarnVersionInfo.getVersion();
        }
        if ((nodeManagerVersion == null) || (VersionUtil.compareVersions(nodeManagerVersion, minimumNodeManagerVersion)) < 0) {
            String message = "Disallowed NodeManager Version " + nodeManagerVersion + ", is less than the minimum version " + minimumNodeManagerVersion + " sending SHUTDOWN signal to " + "NodeManager.";
            LOG.info(message);
            response.setDiagnosticsMessage(message);
            response.setNodeAction(NodeAction.SHUTDOWN);
            return response;
        }
    }
    // Check if this node is a 'valid' node
    if (!this.nodesListManager.isValidNode(host) && !isNodeInDecommissioning(nodeId)) {
        String message = "Disallowed NodeManager from  " + host + ", Sending SHUTDOWN signal to the NodeManager.";
        LOG.info(message);
        response.setDiagnosticsMessage(message);
        response.setNodeAction(NodeAction.SHUTDOWN);
        return response;
    }
    // check if node's capacity is load from dynamic-resources.xml
    String nid = nodeId.toString();
    Resource dynamicLoadCapability = loadNodeResourceFromDRConfiguration(nid);
    if (dynamicLoadCapability != null) {
        if (LOG.isDebugEnabled()) {
            LOG.debug("Resource for node: " + nid + " is adjusted from: " + capability + " to: " + dynamicLoadCapability + " due to settings in dynamic-resources.xml.");
        }
        capability = dynamicLoadCapability;
        // sync back with new resource.
        response.setResource(capability);
    }
    // Check if this node has minimum allocations
    if (capability.getMemorySize() < minAllocMb || capability.getVirtualCores() < minAllocVcores) {
        String message = "NodeManager from  " + host + " doesn't satisfy minimum allocations, Sending SHUTDOWN" + " signal to the NodeManager.";
        LOG.info(message);
        response.setDiagnosticsMessage(message);
        response.setNodeAction(NodeAction.SHUTDOWN);
        return response;
    }
    response.setContainerTokenMasterKey(containerTokenSecretManager.getCurrentKey());
    response.setNMTokenMasterKey(nmTokenSecretManager.getCurrentKey());
    RMNode rmNode = new RMNodeImpl(nodeId, rmContext, host, cmPort, httpPort, resolve(host), capability, nodeManagerVersion, physicalResource);
    RMNode oldNode = this.rmContext.getRMNodes().putIfAbsent(nodeId, rmNode);
    if (oldNode == null) {
        this.rmContext.getDispatcher().getEventHandler().handle(new RMNodeStartedEvent(nodeId, request.getNMContainerStatuses(), request.getRunningApplications()));
    } else {
        LOG.info("Reconnect from the node at: " + host);
        this.nmLivelinessMonitor.unregister(nodeId);
        // Reset heartbeat ID since node just restarted.
        oldNode.resetLastNodeHeartBeatResponse();
        this.rmContext.getDispatcher().getEventHandler().handle(new RMNodeReconnectEvent(nodeId, rmNode, request.getRunningApplications(), request.getNMContainerStatuses()));
    }
    // On every node manager register we will be clearing NMToken keys if
    // present for any running application.
    this.nmTokenSecretManager.removeNodeKey(nodeId);
    this.nmLivelinessMonitor.register(nodeId);
    // RMNode inserted
    if (!rmContext.isWorkPreservingRecoveryEnabled()) {
        if (!request.getNMContainerStatuses().isEmpty()) {
            LOG.info("received container statuses on node manager register :" + request.getNMContainerStatuses());
            for (NMContainerStatus status : request.getNMContainerStatuses()) {
                handleNMContainerStatus(status, nodeId);
            }
        }
    }
    // Update node's labels to RM's NodeLabelManager.
    Set<String> nodeLabels = NodeLabelsUtils.convertToStringSet(request.getNodeLabels());
    if (isDistributedNodeLabelsConf && nodeLabels != null) {
        try {
            updateNodeLabelsFromNMReport(nodeLabels, nodeId);
            response.setAreNodeLabelsAcceptedByRM(true);
        } catch (IOException ex) {
            // Ensure the exception is captured in the response
            response.setDiagnosticsMessage(ex.getMessage());
            response.setAreNodeLabelsAcceptedByRM(false);
        }
    } else if (isDelegatedCentralizedNodeLabelsConf) {
        this.rmContext.getRMDelegatedNodeLabelsUpdater().updateNodeLabels(nodeId);
    }
    StringBuilder message = new StringBuilder();
    message.append("NodeManager from node ").append(host).append("(cmPort: ").append(cmPort).append(" httpPort: ");
    message.append(httpPort).append(") ").append("registered with capability: ").append(capability);
    message.append(", assigned nodeId ").append(nodeId);
    if (response.getAreNodeLabelsAcceptedByRM()) {
        message.append(", node labels { ").append(StringUtils.join(",", nodeLabels) + " } ");
    }
    LOG.info(message.toString());
    response.setNodeAction(NodeAction.NORMAL);
    response.setRMIdentifier(ResourceManager.getClusterTimeStamp());
    response.setRMVersion(YarnVersionInfo.getVersion());
    return response;
}

Example 2 with RMNodeReconnectEvent

use of org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent in project hadoop by apache.

the class TestRMNodeTransitions method testReconnnectUpdate.

@Test
public void testReconnnectUpdate() {
    final String nmVersion1 = "nm version 1";
    final String nmVersion2 = "nm version 2";
    RMNodeImpl node = getRunningNode(nmVersion1);
    Assert.assertEquals(nmVersion1, node.getNodeManagerVersion());
    RMNodeImpl reconnectingNode = getRunningNode(nmVersion2);
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), reconnectingNode, null, null));
    Assert.assertEquals(nmVersion2, node.getNodeManagerVersion());
}

Example 3 with RMNodeReconnectEvent

use of org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent in project hadoop by apache.

the class TestRMNodeTransitions method testReconnectOnDecommissioningNode.

@Test
public void testReconnectOnDecommissioningNode() {
    RMNodeImpl node = getDecommissioningNode();
    ClusterMetrics cm = ClusterMetrics.getMetrics();
    int initialActive = cm.getNumActiveNMs();
    int initialDecommissioning = cm.getNumDecommissioningNMs();
    int initialDecommissioned = cm.getNumDecommisionedNMs();
    // Reconnect event with running app
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), node, getAppIdList(), null));
    // still decommissioning
    Assert.assertEquals(NodeState.DECOMMISSIONING, node.getState());
    Assert.assertEquals("Active Nodes", initialActive, cm.getNumActiveNMs());
    Assert.assertEquals("Decommissioning Nodes", initialDecommissioning, cm.getNumDecommissioningNMs());
    Assert.assertEquals("Decommissioned Nodes", initialDecommissioned, cm.getNumDecommisionedNMs());
    // Reconnect event without any running app
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), node, null, null));
    Assert.assertEquals(NodeState.DECOMMISSIONED, node.getState());
    Assert.assertEquals("Active Nodes", initialActive, cm.getNumActiveNMs());
    Assert.assertEquals("Decommissioning Nodes", initialDecommissioning - 1, cm.getNumDecommissioningNMs());
    Assert.assertEquals("Decommissioned Nodes", initialDecommissioned + 1, cm.getNumDecommisionedNMs());
}

Example 4 with RMNodeReconnectEvent

use of org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent in project hadoop by apache.

the class TestRMNodeTransitions method testReconnect.

@Test
public void testReconnect() {
    RMNodeImpl node = getRunningNode();
    ClusterMetrics cm = ClusterMetrics.getMetrics();
    int initialActive = cm.getNumActiveNMs();
    int initialLost = cm.getNumLostNMs();
    int initialUnhealthy = cm.getUnhealthyNMs();
    int initialDecommissioned = cm.getNumDecommisionedNMs();
    int initialRebooted = cm.getNumRebootedNMs();
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), node, null, null));
    Assert.assertEquals("Active Nodes", initialActive, cm.getNumActiveNMs());
    Assert.assertEquals("Lost Nodes", initialLost, cm.getNumLostNMs());
    Assert.assertEquals("Unhealthy Nodes", initialUnhealthy, cm.getUnhealthyNMs());
    Assert.assertEquals("Decommissioned Nodes", initialDecommissioned, cm.getNumDecommisionedNMs());
    Assert.assertEquals("Rebooted Nodes", initialRebooted, cm.getNumRebootedNMs());
    Assert.assertEquals(NodeState.RUNNING, node.getState());
    Assert.assertNotNull(nodesListManagerEvent);
    Assert.assertEquals(NodesListManagerEventType.NODE_USABLE, nodesListManagerEvent.getType());
}

Example 5 with RMNodeReconnectEvent

use of org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeReconnectEvent in project hadoop by apache.

the class TestRMNodeTransitions method testReconnectWithNewPortOnDecommissioningNode.

@Test
public void testReconnectWithNewPortOnDecommissioningNode() {
    RMNodeImpl node = getDecommissioningNode();
    Random r = new Random();
    node.setHttpPort(r.nextInt(10000));
    // Reconnect event with running app
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), node, getAppIdList(), null));
    // still decommissioning
    Assert.assertEquals(NodeState.DECOMMISSIONING, node.getState());
    node.setHttpPort(r.nextInt(10000));
    // Reconnect event without any running app
    node.handle(new RMNodeReconnectEvent(node.getNodeID(), node, null, null));
    Assert.assertEquals(NodeState.DECOMMISSIONED, node.getState());
}