com.amazonaws.services.elasticmapreduce.model.Tag Java Examples

The following examples show how to use com.amazonaws.services.elasticmapreduce.model.Tag. You can vote up the ones you like or vote down the ones you don't like, and go to the original project or source file by following the links above each example. You may check out the related API usage on the sidebar.
Example #1
Source File: EmrClusterTableProviderTest.java    From aws-athena-query-federation with Apache License 2.0 6 votes vote down vote up
private Cluster makeCluster(String id)
{
    return new Cluster()
            .withId(id)
            .withName("name")
            .withAutoScalingRole("autoscaling_role")
            .withCustomAmiId("custom_ami")
            .withInstanceCollectionType("instance_collection_type")
            .withLogUri("log_uri")
            .withMasterPublicDnsName("master_public_dns")
            .withReleaseLabel("release_label")
            .withRunningAmiVersion("running_ami")
            .withScaleDownBehavior("scale_down_behavior")
            .withServiceRole("service_role")
            .withApplications(new Application().withName("name").withVersion("version"))
            .withTags(new Tag("key", "value"));
}
 
Example #2
Source File: EmrDaoImpl.java    From herd with Apache License 2.0 6 votes vote down vote up
/**
 * Create the tag list for the EMR nodes.
 *
 * @param emrClusterDefinition the EMR definition name value.
 *
 * @return list of all tag definitions for the given configuration.
 */
private List<Tag> getEmrTags(EmrClusterDefinition emrClusterDefinition)
{
    List<Tag> tags = new ArrayList<>();

    // Get the nodeTags from xml
    for (NodeTag thisTag : emrClusterDefinition.getNodeTags())
    {
        // Create a AWS tag and add
        if (StringUtils.isNotBlank(thisTag.getTagName()) && StringUtils.isNotBlank(thisTag.getTagValue()))
        {
            tags.add(new Tag(thisTag.getTagName(), thisTag.getTagValue()));
        }
    }

    // Return the object
    return tags;
}
 
Example #3
Source File: LambdaContainer.java    From aws-big-data-blog with Apache License 2.0 6 votes vote down vote up
protected List<String> getActiveTaggedClusters() throws Exception{
	AmazonElasticMapReduceClient emrClient = new AmazonElasticMapReduceClient();
	List<String> waitingClusters = new ArrayList<String>();
	ListClustersResult clusterResult = emrClient.listClusters(new ListClustersRequest().withClusterStates(ClusterState.WAITING));
	
	DescribeClusterRequest specifcTagDescribe = new DescribeClusterRequest();
	specifcTagDescribe.putCustomQueryParameter("Cluster.Tags",null);
	 for( ClusterSummary cluster : clusterResult.getClusters()){
		 	System.out.println("list cluster id "+cluster.getId());
		 	List<Tag> tagList = emrClient.describeCluster(specifcTagDescribe.withClusterId(cluster.getId())).getCluster().getTags();
		 	for(Tag tag:tagList){
		 		if(tag.getKey().equals(props.getProperty("edba.cluster.tag.key"))){
		 			waitingClusters.add(cluster.getId());
		 		}
		 	}
		 	
	}
	return waitingClusters;
	
}
 
Example #4
Source File: EmrDaoTest.java    From herd with Apache License 2.0 4 votes vote down vote up
@Test
public void createEmrClusterAssertCallRunEmrJobFlowWithInstanceFleetAndMultipleSubnets()
{
    // Create objects required for testing.
    final String clusterName = "clusterName";
    final String clusterId = "clusterId";
    final String name = STRING_VALUE;
    final String instanceFleetType = STRING_VALUE_2;
    final Integer targetOnDemandCapacity = INTEGER_VALUE;
    final Integer targetSpotCapacity = INTEGER_VALUE_2;
    final List<EmrClusterDefinitionInstanceTypeConfig> emrClusterDefinitionInstanceTypeConfigs = null;
    final EmrClusterDefinitionLaunchSpecifications emrClusterDefinitionLaunchSpecifications = null;
    final EmrClusterDefinitionInstanceFleet emrClusterDefinitionInstanceFleet =
        new EmrClusterDefinitionInstanceFleet(name, instanceFleetType, targetOnDemandCapacity, targetSpotCapacity, emrClusterDefinitionInstanceTypeConfigs,
            emrClusterDefinitionLaunchSpecifications);

    // Create an EMR cluster definition with instance fleet configuration and multiple EC2 subnet IDs.
    EmrClusterDefinition emrClusterDefinition = new EmrClusterDefinition();
    emrClusterDefinition.setInstanceFleets(Lists.newArrayList(emrClusterDefinitionInstanceFleet));
    emrClusterDefinition.setSubnetId(String.format("%s , %s  ", EC2_SUBNET, EC2_SUBNET_2));
    emrClusterDefinition.setNodeTags(Lists.newArrayList(new NodeTag("tagName", "tagValue")));

    when(mockEmrOperations.runEmrJobFlow(any(), any())).then(new Answer<String>()
    {
        @Override
        public String answer(InvocationOnMock invocation)
        {
            // Assert that the given EMR cluster definition produced the correct RunJobFlowRequest.
            RunJobFlowRequest runJobFlowRequest = invocation.getArgument(1);
            JobFlowInstancesConfig jobFlowInstancesConfig = runJobFlowRequest.getInstances();
            assertEquals(0, CollectionUtils.size(jobFlowInstancesConfig.getInstanceGroups()));
            final List<InstanceTypeConfig> expectedInstanceTypeConfigs = null;
            assertEquals(Lists.newArrayList(
                new InstanceFleetConfig().withName(name).withInstanceFleetType(instanceFleetType).withTargetOnDemandCapacity(targetOnDemandCapacity)
                    .withTargetSpotCapacity(targetSpotCapacity).withInstanceTypeConfigs(expectedInstanceTypeConfigs).withLaunchSpecifications(null)),
                jobFlowInstancesConfig.getInstanceFleets());
            assertNull(jobFlowInstancesConfig.getEc2SubnetId());
            assertEquals(2, CollectionUtils.size(jobFlowInstancesConfig.getEc2SubnetIds()));
            assertTrue(jobFlowInstancesConfig.getEc2SubnetIds().contains(EC2_SUBNET));
            assertTrue(jobFlowInstancesConfig.getEc2SubnetIds().contains(EC2_SUBNET_2));
            assertEquals(herdStringHelper.getRequiredConfigurationValue(ConfigurationValue.EMR_DEFAULT_EC2_NODE_IAM_PROFILE_NAME),
                runJobFlowRequest.getJobFlowRole());
            assertEquals(herdStringHelper.getRequiredConfigurationValue(ConfigurationValue.EMR_DEFAULT_SERVICE_IAM_ROLE_NAME),
                runJobFlowRequest.getServiceRole());
            List<StepConfig> stepConfigs = runJobFlowRequest.getSteps();
            assertEquals(0, stepConfigs.size());
            List<Tag> tags = runJobFlowRequest.getTags();
            assertEquals(1, tags.size());
            {
                Tag tag = tags.get(0);
                assertEquals("tagName", tag.getKey());
                assertEquals("tagValue", tag.getValue());
            }

            return clusterId;
        }
    });

    assertEquals(clusterId, emrDao.createEmrCluster(clusterName, emrClusterDefinition, getAwsParamsDto()));
}
 
Example #5
Source File: EmrOperatorFactory.java    From digdag with Apache License 2.0 4 votes vote down vote up
private NewCluster submitNewClusterRequest(AmazonElasticMapReduce emr, String tag, StepCompiler stepCompiler,
        Config cluster, Filer filer, ParameterCompiler parameterCompiler)
        throws IOException
{
    RemoteFile runner = prepareRunner(filer, tag);

    // Compile steps
    stepCompiler.compile(runner);

    List<StepConfig> stepConfigs = stepCompiler.stepConfigs();

    Config ec2 = cluster.getNested("ec2");
    Config master = ec2.getNestedOrGetEmpty("master");
    List<Config> core = ec2.getOptional("core", Config.class).transform(ImmutableList::of).or(ImmutableList.of());
    List<Config> task = ec2.getListOrEmpty("task", Config.class);

    List<String> applications = cluster.getListOrEmpty("applications", String.class);
    if (applications.isEmpty()) {
        applications = ImmutableList.of("Hadoop", "Hive", "Spark", "Flink");
    }

    // TODO: allow configuring additional application parameters
    List<Application> applicationConfigs = applications.stream()
            .map(application -> new Application().withName(application))
            .collect(toList());

    // TODO: merge configurations with the same classification?
    List<Configuration> configurations = cluster.getListOrEmpty("configurations", JsonNode.class).stream()
            .map(this::configurations)
            .flatMap(Collection::stream)
            .collect(toList());

    List<JsonNode> bootstrap = cluster.getListOrEmpty("bootstrap", JsonNode.class);
    List<BootstrapActionConfig> bootstrapActions = new ArrayList<>();
    for (int i = 0; i < bootstrap.size(); i++) {
        bootstrapActions.add(bootstrapAction(i + 1, bootstrap.get(i), tag, filer, runner, parameterCompiler));
    }

    // Stage files to S3
    filer.stageFiles();

    Optional<String> subnetId = ec2.getOptional("subnet_id", String.class);

    String defaultMasterInstanceType;
    String defaultCoreInstanceType;
    String defaultTaskInstanceType;

    if (subnetId.isPresent()) {
        // m4 requires VPC (subnet id)
        defaultMasterInstanceType = "m4.2xlarge";
        defaultCoreInstanceType = "m4.xlarge";
        defaultTaskInstanceType = "m4.xlarge";
    }
    else {
        defaultMasterInstanceType = "m3.2xlarge";
        defaultCoreInstanceType = "m3.xlarge";
        defaultTaskInstanceType = "m3.xlarge";
    }

    RunJobFlowRequest request = new RunJobFlowRequest()
            .withName(cluster.get("name", String.class, "Digdag") + " (" + tag + ")")
            .withReleaseLabel(cluster.get("release", String.class, "emr-5.2.0"))
            .withSteps(stepConfigs)
            .withBootstrapActions(bootstrapActions)
            .withApplications(applicationConfigs)
            .withLogUri(cluster.get("logs", String.class, null))
            .withJobFlowRole(cluster.get("cluster_role", String.class, "EMR_EC2_DefaultRole"))
            .withServiceRole(cluster.get("service_role", String.class, "EMR_DefaultRole"))
            .withTags(new Tag().withKey("DIGDAG_CLUSTER_ID").withValue(tag))
            .withVisibleToAllUsers(cluster.get("visible", boolean.class, true))
            .withConfigurations(configurations)
            .withInstances(new JobFlowInstancesConfig()
                    .withInstanceGroups(ImmutableList.<InstanceGroupConfig>builder()
                            // Master Node
                            .add(instanceGroupConfig("Master", master, "MASTER", defaultMasterInstanceType, 1))
                            // Core Group
                            .addAll(instanceGroupConfigs("Core", core, "CORE", defaultCoreInstanceType))
                            // Task Groups
                            .addAll(instanceGroupConfigs("Task %d", task, "TASK", defaultTaskInstanceType))
                            .build()
                    )
                    .withAdditionalMasterSecurityGroups(ec2.getListOrEmpty("additional_master_security_groups", String.class))
                    .withAdditionalSlaveSecurityGroups(ec2.getListOrEmpty("additional_slave_security_groups", String.class))
                    .withEmrManagedMasterSecurityGroup(ec2.get("emr_managed_master_security_group", String.class, null))
                    .withEmrManagedSlaveSecurityGroup(ec2.get("emr_managed_slave_security_group", String.class, null))
                    .withServiceAccessSecurityGroup(ec2.get("service_access_security_group", String.class, null))
                    .withTerminationProtected(cluster.get("termination_protected", boolean.class, false))
                    .withPlacement(cluster.getOptional("availability_zone", String.class)
                            .transform(zone -> new PlacementType().withAvailabilityZone(zone)).orNull())
                    .withEc2SubnetId(subnetId.orNull())
                    .withEc2KeyName(ec2.get("key", String.class))
                    .withKeepJobFlowAliveWhenNoSteps(!cluster.get("auto_terminate", boolean.class, true)));

    logger.info("Submitting EMR job with {} steps(s)", request.getSteps().size());
    RunJobFlowResult result = emr.runJobFlow(request);
    logger.info("Submitted EMR job with {} step(s): {}", request.getSteps().size(), result.getJobFlowId(), result);

    return NewCluster.of(result.getJobFlowId(), request.getSteps().size());
}