.getResponseId()) { LOG.info("Received duplicate heartbeat from node " + rmNode.getNodeAddress()+ " responseId=" + remoteNodeStatus.getResponseId()); return lastNodeHeartbeatResponse; } else if (remoteNodeStatus.getResponseId() + 1 < lastNodeHeartbeatResponse .getResponseId()) { String message = "Too far behind rm response id:" + lastNodeHeartbeatResponse.getResponseId() + " nm response id:" + remoteNodeStatus.getResponseId(); LOG.info(message); getResponseId() + 1, NodeAction.NORMAL, null, null, null, null, nextHeartBeatInterval); rmNode.updateNodeHeartbeatResponseForCleanup(nodeHeartBeatResponse);
.getResponseId()) { LOG.info("Received duplicate heartbeat from node " + rmNode.getNodeAddress()+ " responseId=" + remoteNodeStatus.getResponseId()); return lastNodeHeartbeatResponse; } else if (remoteNodeStatus.getResponseId() + 1 < lastNodeHeartbeatResponse .getResponseId()) { String message = "Too far behind rm response id:" + lastNodeHeartbeatResponse.getResponseId() + " nm response id:" + remoteNodeStatus.getResponseId(); LOG.info(message); getResponseId() + 1, NodeAction.NORMAL, null, null, null, null, nextHeartBeatInterval); rmNode.updateNodeHeartbeatResponseForCleanup(nodeHeartBeatResponse);
public void heartbeat() throws IOException, YarnException { NodeStatus nodeStatus = org.apache.hadoop.yarn.server.resourcemanager.NodeManager.createNodeStatus( nodeId, getContainerStatuses(containers)); nodeStatus.setResponseId(responseID); NodeHeartbeatRequest request = recordFactory .newRecordInstance(NodeHeartbeatRequest.class); request.setNodeStatus(nodeStatus); NodeHeartbeatResponse response = resourceTrackerService .nodeHeartbeat(request); responseID = response.getResponseId(); }
public void heartbeat() throws IOException, YarnException { NodeStatus nodeStatus = org.apache.hadoop.yarn.server.resourcemanager.NodeManager.createNodeStatus( nodeId, getContainerStatuses(containers)); nodeStatus.setResponseId(responseID); NodeHeartbeatRequest request = recordFactory .newRecordInstance(NodeHeartbeatRequest.class); request.setNodeStatus(nodeStatus); NodeHeartbeatResponse response = resourceTrackerService .nodeHeartbeat(request); responseID = response.getResponseId(); }
.getContainersToBeRemovedFromNM()); lastHeartBeatID = response.getResponseId(); List<ContainerId> containersToCleanup = response .getContainersToCleanup();
if (getNextResponseId( remoteNodeStatus.getResponseId()) == lastNodeHeartbeatResponse .getResponseId()) { LOG.info("Received duplicate heartbeat from node " + rmNode.getNodeAddress()+ " responseId=" + remoteNodeStatus.getResponseId()); return lastNodeHeartbeatResponse; } else if (remoteNodeStatus.getResponseId() != lastNodeHeartbeatResponse .getResponseId()) { String message = "Too far behind rm response id:" + lastNodeHeartbeatResponse.getResponseId() + " nm response id:" + remoteNodeStatus.getResponseId(); LOG.info(message); getNextResponseId(lastNodeHeartbeatResponse.getResponseId()), NodeAction.NORMAL, null, null, null, null, nextHeartBeatInterval); rmNode.setAndUpdateNodeHeartbeatResponse(nodeHeartBeatResponse);
.getContainersToBeRemovedFromNM()); lastHeartBeatID = response.getResponseId(); List<ContainerId> containersToCleanup = response .getContainersToCleanup();
lastHeartbeatID = response.getResponseId(); List<ContainerId> containersToCleanup = response .getContainersToCleanup();
public void run() { int lastResponseID = 0; while (!stopT) { try { org.apache.hadoop.yarn.server.api.records.NodeStatus nodeStatus = recordFactory .newRecordInstance(org.apache.hadoop.yarn.server.api.records.NodeStatus.class); nodeStatus.setNodeId(request3.getNodeId()); nodeStatus.setResponseId(lastResponseID); nodeStatus.setNodeHealthStatus(recordFactory.newRecordInstance(NodeHealthStatus.class)); nodeStatus.getNodeHealthStatus().setIsNodeHealthy(true); NodeHeartbeatRequest request = recordFactory .newRecordInstance(NodeHeartbeatRequest.class); request.setNodeStatus(nodeStatus); lastResponseID = resourceTrackerService.nodeHeartbeat(request) .getResponseId(); Thread.sleep(1000); } catch(Exception e) { LOG.info("failed to heartbeat ", e); } } } }
public void run() { int lastResponseID = 0; while (!stopT) { try { org.apache.hadoop.yarn.server.api.records.NodeStatus nodeStatus = recordFactory .newRecordInstance(org.apache.hadoop.yarn.server.api.records.NodeStatus.class); nodeStatus.setNodeId(request3.getNodeId()); nodeStatus.setResponseId(lastResponseID); nodeStatus.setNodeHealthStatus(recordFactory.newRecordInstance(NodeHealthStatus.class)); nodeStatus.getNodeHealthStatus().setIsNodeHealthy(true); NodeHeartbeatRequest request = recordFactory .newRecordInstance(NodeHeartbeatRequest.class); request.setNodeStatus(nodeStatus); lastResponseID = resourceTrackerService.nodeHeartbeat(request) .getResponseId(); Thread.sleep(1000); } catch(Exception e) { LOG.info("failed to heartbeat ", e); } } } }
lastHeartbeatID = response.getResponseId(); List<ContainerId> containersToCleanup = response .getContainersToCleanup();
@Test public void testResponseIdOverflow() throws Exception { Configuration conf = new Configuration(); rm = new MockRM(conf); rm.start(); MockNM nm1 = rm.registerNode("host1:1234", 5120); NodeHeartbeatResponse nodeHeartbeat = nm1.nodeHeartbeat(true); Assert.assertEquals(NodeAction.NORMAL, nodeHeartbeat.getNodeAction()); // prepare the responseId that's about to overflow RMNode node = rm.getRMContext().getRMNodes().get(nm1.getNodeId()); node.getLastNodeHeartBeatResponse().setResponseId(Integer.MAX_VALUE); nm1.setResponseId(Integer.MAX_VALUE); // heartbeat twice and check responseId nodeHeartbeat = nm1.nodeHeartbeat(true); Assert.assertEquals(NodeAction.NORMAL, nodeHeartbeat.getNodeAction()); Assert.assertEquals(0, nodeHeartbeat.getResponseId()); nodeHeartbeat = nm1.nodeHeartbeat(true); Assert.assertEquals(NodeAction.NORMAL, nodeHeartbeat.getNodeAction()); Assert.assertEquals(1, nodeHeartbeat.getResponseId()); } }
responseId = heartbeatResponse.getResponseId();
NodeHeartbeatResponse response = resourceTrackerService.nodeHeartbeat( nodeHeartBeatRequest); Assert.assertTrue(response.getResponseId() == 1); nodeStatus.setResponseId(response.getResponseId()); response = resourceTrackerService.nodeHeartbeat(nodeHeartBeatRequest); Assert.assertTrue(response.getResponseId() == 2); Assert.assertTrue(response.getResponseId() == 2);
NodeHeartbeatResponse response = resourceTrackerService.nodeHeartbeat( nodeHeartBeatRequest); Assert.assertTrue(response.getResponseId() == 1); nodeStatus.setResponseId(response.getResponseId()); response = resourceTrackerService.nodeHeartbeat(nodeHeartBeatRequest); Assert.assertTrue(response.getResponseId() == 2); Assert.assertTrue(response.getResponseId() == 2);
.getResponseId() > 0) { Thread.sleep(WAIT_SLEEP_MS);
.getResponseId() > 0) { Thread.sleep(WAIT_SLEEP_MS);