Home

Cloudera Distribution of Apache Hadoop (CDH)

Before installing, ensure to check and complete the installation requirements. Follow the below instructions to download, install, and set up Unravel for the CDH platform.

Notice

The following instructions are for a single cluster environment. For installing Unravel on a multi-cluster environment, refer to Multi-cluster install.

1. Download Unravel
2. Deploy Unravel binaries

Unravel binaries are available as a tar file or RPM package. You can deploy the Unravel binaries in any directory on the server. However, the user who installs Unravel must have the write permissions to the directory where the Unravel binaries are deployed.

If the binaries are deployed to <Unravel_installation_directory>, Unravel will be available in <Unravel_installation_directory>/unravel. The directory layout for both the Tar and RPM will be unravel/versions/<Directories and files>.

Option 1: Deploy Unravel from a tar file

The following steps to deploy Unravel from a tar file must be performed by a user, who will run Unravel.

  1. Create an Installation directory.

    mkdir /path/to/installation/directory

    For example: mkdir /opt/unravel

    Note

    Some locations may require root access to create a directory. In such a case, after the directory is created, change the ownership to unravel user and continue with the installation procedure as the unravel user. 

    chown -R username:groupname /path/to/installation/directory

    For example: chown -R unravel:unravelgroup /opt/unravel

  2. Extract and copy the Unravel tar file to the installation directory, which was created in the first step. After you extract the contents of the tar file, unravel directory is created within the installation directory.

    tar -zxf unravel-<version>tar.gz -C /<Unravel-installation-directory>

    For example: tar -zxf unravel-4.7.x.x.tar.gz -C /opt The Unravel directory will be available within /opt

Option 2: Deploy Unravel from an RPM package

Important

The following steps, to deploy Unravel from an RPM package, should be performed by a root user. After the RPM package is deployed, the remaining installation procedures should be performed by the unravel user.

  1. Create an installation directory.

    mkdir /usr/local/unravel

  2. Grant ownership of the directory to a user who runs Unravel. This user executes all the processes involved in Unravel installation.

    chown -R username:groupname /user/local/unravel

    For example: chown -R unravel:unravelgroup /usr/local/unravel The Unravel directory is available in /user/local.

  3. Run the following command:

    rpm -i unravel-<version>.rpm

    For example: rpm -i unravel-4.7.x.x.rpm

    In case you want to provide a different location, you can do so by using the --prefix command. For example:

    mkdir /opt/unravel
chown -R username:groupname /opt/unravel
rpm -i unravel-4.7.0.0.rpm --prefix /opt

    The Unravel directory is available in /opt.

  4. Continue with the installation procedures as Unravel user.

3. Run the setup

You can run the setup command to install Unravel. The setup command allows you to do the following:

  • Runs Precheck automatically to detect possible issues that prevent a successful installation. Suggestions are provided to resolve issues. Refer to Precheck filters for the expected value for each filter.

  • Let you run extra parameters to integrate the database of your choice.

    The setup command allows you to use a managed database shipped with Unravel or an external database. The setup uses the Unravel managed PostgreSQL database when run without any additional parameters. Otherwise, you can specify one of the following types of databases in the setup command:

    • MySQL (Unravel managed as well as external MySQL database)

    • MariaDB (Unravel managed as well as external MariaDB database)

    • PostgreSQL (External PostgreSQL)

    Refer to Integrate database for details.

  • Let you specify a separate path for the data directory other than the default path.

    You can locate the Unravel data and configurations in the data directory. By default, the installer maintains the data directory under <Unravel installation directory>/data. You can also change the data directory's default location by running additional parameters with the setup command. To install Unravel with the setup command.

  • Provides more setup options.

Notice

The Unravel user who owns the installation directory should run the setup command to install Unravel.

To install Unravel with the setup command, do the following:

  1. Switch to Unravel user.

      su - <unravel user>

  2. Run setup command:

    Note

    Refer to setup Options for all the additional parameters that you can run with the setup command

    Refer to Integrate database topic and complete the pre-requisites before running the setup command with any other database other than Unravel managed PostgreSQL, which is shipped with the product. Extra parameters must be passed with the setup command when using another database.

    Tip

    Optionally, if you want to provide a different data directory, you can pass an extra parameter (--data-directory) with the setup command as follows:

    <unravel_installation_directory>/unravel/versions/<Unravel version>/setup --data-directory /the/data/directory

    Similarly, you can configure separate directories for other unravel directories —contact support for assistance.

    • PostgreSQL

      • Unravel managed PostgreSQL

        <unravel_installation_directory>/unravel/versions/<Unravel version>/setup

      • External PostgreSQL

        <unravel_installation_directory>/unravel/versions/<Unravel version>/setup --external-database postgresql <HOST> <PORT> <SCHEMA> <USERNAME> <PASSWORD>

        The HOST, PORT, SCHEMA, USERNAME, and PASSWORD are optional fields and are prompted if missing. For example: /opt/unravel/versions/abcd.992/setup --external-database postgresql xyz.unraveldata.com 5432 unravel_db_prod unravel unraveldata

    • MySQL

      • Unravel managed MySQL

        <unravel_installation_directory>/unravel/versions/<Unravel version>/setup --extra /tmp/mysql

      • External MySQL

        <unravel_installation_directory>/unravel/versions/<Unravel version>/setup --extra /tmp/<MySQL-directory> --external-database mysql <HOST> <PORT> <SCHEMA> <USERNAME> <PASSWORD>

        The HOST, PORT, SCHEMA, USERNAME, and PASSWORD are optional fields and are prompted if missing.

    • MariaDB

      • Unravel managed MariaDB

        <unravel_installation_directory>/unravel/versions/<Unravel version>/setup --extra /tmp/mariadb

      • External MariaDB

        <unravel_installation_directory>unravel/versions/<Unravel version>/setup --extra /tmp/<MariaDB-directory> --external-database mariadb <HOST> <PORT> <SCHEMA> <USERNAME> <PASSWORD>

        The HOST, PORT, SCHEMA, USERNAME, and PASSWORD are optional fields and are prompted if missing.

    Precheck is automatically run when you run the setup command. Refer to Precheck filters for the expected value for each filter.

  3. Apply the changes.

    <Unravel installation directory>/unravel/manager config apply

  4. Start all the services.

    <unravel_installation_directory>/unravel/manager start

  5. Check the status of services.

    <unravel_installation_directory>/unravel/manager report

    The following service statuses are reported:

    • OK: Service is up and running.

    • Not Monitored: Service is not running. (Has stopped or has failed to start)

    • Initializing: Services are starting up.

    • Does not exist: The process unexpectedly disappeared. A restart will be attempted ten times.

    You can also get the status and information for a specific service. Run the manager report command as follows:

    <unravel_installation_directory>/unravel/manager report <service>

    For example: /opt/unravel/manager report auto_action

Precheck

The Precheck output displays the issues that prevent a successful installation and provides suggestions to resolve them. You must resolve each of the issues before proceeding. See Precheck filters.

After resolving the precheck issues, you must re-login or reload the shell to execute the setup command again.

Note

You can skip the precheck using the setup --skip-precheck command in certain situations.

For example:

/opt/unravel/versions/<Unravel version>/setup --skip-precheck

You can also skip the checks that you know can fail. For example, if you want to skip the Check limits option and the Disk freespace option, pick the command within the parenthesis corresponding to these failed options and run the setup command as follows:

setup --filter-precheck ~check_limits,~check_freespace

Tip

Run --help with the setup command and any combination of the setup command for complete usage details.

<unravel_installation_directory>/unravel/versions/<Unravel version>/setup --help
Precheck filters
Setup Options
Precheck Sample
/opt/unravel/versions/abcd.1004/setup 
2021-04-05 15:51:30 Sending logs to: /tmp/unravel-setup-20210405-155130.log
2021-04-05 15:51:30 Running preinstallation check...
2021-04-05 15:51:31 Gathering information ................. Ok
2021-04-05 15:51:51 Running checks .................. Ok
--------------------------------------------------------------------------------
system
 Check limits        : PASSED
 Clock sync          : PASSED
 CPU requirement     : PASSED, Available cores: 8 cores
 Disk access         : PASSED, /opt/unravel/versions/develop.1004/healthcheck/healthcheck/plugins/system is writable
 Disk freespace      : PASSED, 229 GB of free disk space is available for precheck dir.
 Kerberos tools      : PASSED
 Memory requirement  : PASSED, Available memory: 79 GB
 Network ports       : PASSED
 OS libraries        : PASSED
 OS release          : PASSED, OS release version: centos 7.6
 OS settings         : PASSED
 SELinux             : PASSED
--------------------------------------------------------------------------------
Healthcheck report bundle: /tmp/healthcheck-20210405155130-xyz.unraveldata.com.tar.gz
2021-04-05 15:51:53 Prepare to install with: /opt/unravel/versions/abcd.1004/installer/installer/../installer/conf/presets/default.yaml
2021-04-05 15:51:57 Sending logs to: /opt/unravel/logs/setup.log
2021-04-05 15:51:57 Instantiating templates ................................................................................................................................................................................................................................ Ok
2021-04-05 15:52:05 Creating parcels .................................... Ok
2021-04-05 15:52:20 Installing sensors file ............................ Ok
2021-04-05 15:52:20 Installing pgsql connector ... Ok
2021-04-05 15:52:22 Starting service monitor ... Ok
2021-04-05 15:52:27 Request start for elasticsearch_1 .... Ok
2021-04-05 15:52:27 Waiting for elasticsearch_1 for 120 sec ......... Ok
2021-04-05 15:52:35 Request start for zookeeper .... Ok
2021-04-05 15:52:35 Request start for kafka .... Ok
2021-04-05 15:52:35 Waiting for kafka for 120 sec ...... Ok
2021-04-05 15:52:37 Waiting for kafka to be alive for 120 sec ..... Ok
2021-04-05 15:52:42 Initializing pgsql ... Ok
2021-04-05 15:52:46 Request start for pgsql .... Ok
2021-04-05 15:52:46 Waiting for pgsql for 120 sec ..... Ok
2021-04-05 15:52:47 Creating database schema ................. Ok
2021-04-05 15:52:50 Generating hashes .... Ok
2021-04-05 15:52:52 Loading elasticsearch templates ............ Ok
2021-04-05 15:52:55 Creating kafka topics .................... Ok
2021-04-05 15:53:36 Creating schema objects ....................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................... Ok
2021-04-05 15:54:03 Request stop ....................................................... Ok
2021-04-05 15:54:16 Done
[unravel@xyz ~]$
4. Add configurations

  1. Run manager config auto command to automatically pull in all the Hadoop configurations. You will be prompted to provide the location and credentials for Cloudera Manager or Ambari UI.

    <unravel_installation_directory>/unravel/manager config auto

    If there are more than one clusters that are handled by Cloudera Manager or Ambari, you will be prompted to enable the cluster that you want to monitor. Run the following command to enable a cluster. 

    <unravel_installation_directory>/unravel/manager config cluster enable <CLUSTER_KEY>

    Example: /opt/unravel/manager config cluster enable cluster1

    Tip

    Here <CLUSTER_KEY> is the name of the cluster that you want to enable for Unravel monitoring. This can be retrieved from the output shown for the manager config auto command.

  2. The Hive metastore database password can be recovered automatically only for a cluster manager with an administrative account. Otherwise, it must be set manually as follows:

    <Unravel installation directory>/unravel/manager config hive metastore password <CLUSTER_KEY> <HIVE_KEY> <PASSWORD>

    Example: /opt/unravel/manager config hive metastore password cluster1 HIVE p@P@SsWorD

    Tip

    Here <CLUSTER_KEY> is the name of the cluster where you want to set the Hive configurations.

    Also, refer to Connecting to Hive metastore in a single cluster environment.

  3. Optional: Set up Kerberos to authenticate Hadoop services.

    • If you are using Kerberos authentication, set the principal path and keytab, enable Kerberos authentication, and apply the changes.

      <Unravel installation directory>/unravel/manager config kerberos set --keytab </path/to/keytab file> --principal <server@example.com>
<Unravel installation directory>/unravel/manager config kerberos enable
<unravel_installation_directory>/manager config apply

    • If you are using Truststore certificates, run the following steps from the manager tool to add certificates to the Truststore:

      1. Download the certificates to a directory.

      2. Provide permissions to the user, who installs unravel, to access the certificates directory.

        chown -R username:groupname /path/to/certificates/directory

      3. Upload the certificates.

        ## Option 1
<unravel_installation_directory>/unravel/manager config tls trust add </path/to/the/certificate/files

or 

## Option 2
<unravel_installation_directory>/unravel/manager config tls trust add --pem </path/to/the/certificate/files>
<unravel_installation_directory>/unravel/manager config tls trust add --jks </path/to/the/certificate/files>
<unravel_installation_directory>/unravel/manager config tls trust add --pkcs12 </path/to/the/certificate/files>

      4. Enable the Truststore

        <unravel_installation_directory>/unravel/manager config tls trust <enable|disable>
<unravel_installation_directory>/unravel/manager config apply

      5. Verify the connection.

        <unravel_installation_directory>/unravel/manager verify connect <Cluster Manager-host> <Cluster Manager-port>

        For example:

        /opt/unravel/manager verify connect xyz.unraveldata.com 7180
-- Running: verify connect xyz.unraveldata.com 7180
 - Resolved IP: 111.17.4.123
 - Reverse lookup: ('xyz.unraveldata.com', [], ['111.17.4.123'])
 - Connection:   OK
 - TLS:      No
-- OK

    • If you are using TLS protocol, refer to Enabling Transport Layer Security (TLS) for Unravel UI.Enabling Transport Layer Security (TLS) for Unravel UI

  4. Apply changes. 

    <unravel_installation_directory>/unravel/manager config apply

  5. Start all the services.

    <unravel_installation_directory>/unravel/manager start

  6. Check the status of services.

    <unravel_installation_directory>/unravel/manager report

    The following service statuses are reported:

    • OK: Service is up and running

    • Not Monitored: Service is not running. (Has stopped or has failed to start)

    • Initializing: Services are starting up.

    • Does not exist: The process unexpectedly disappeared. Restarts will be attempted 10 times.

    You can also get the status and information for a specific service. Run the manager report command as follows:

    <unravel_installation_directory>/unravel/manager report <service>

    For example: /opt/unravel/manager report auto_action

  7. Enable additional instrumentation. (CDH, CDP)

  8. Set additional configurations, if required.

  9. Optionally, you can run healthcheck, at this point, to verify that all the configurations and services are running successfully.

    <unravel_installation_directory>/unravel/manager healthcheck

    Healthcheck is run automatically on an hourly basis in the backend. You can set your email to receive the healthcheck reports.

5. Enable additional instrumentation for CDHƒ

This topic explains how to enable additional instrumentation on your gateway/edge/client nodes that are used to submit jobs to your big data platform. Additional instrumentation can include:

  • Sensor jars are packaged in a parcel on Unravel server.

  • Hive queries in Hadoop that are pushed to Unravel Server by the Hive Hook sensor, a JAR file.

  • Spark job performance metrics that are pushed to Unravel Server by the Spark sensor, a JAR file.

  • Copying Hive hook and Btrace jars to HDFS shared library path.

  • Impala queries that are pulled from the Cloudera Manager

1. Download, distribute, and activate Unravel sensor

Sensor JARs are packaged in a parcel on Unravel server. Run the following steps from the Cloudera Manager to download, distribute, and activate this parcel.

Note

Ensure that Unravel is up and running before you perform the following steps.

  1. In Cloudera Manager, click package.png. The Parcel page is displayed.

  2. On the Parcel page, click Configuration or Parcel Repositories & Network settings. The Parcel Configurations dialog box is displayed.

  3. Go to the Remote Parcel Repository URLs section, click + and enter the Unravel host along with the exact directory name for your CDH version.

    http://<unravel-host>:<port>/parcels/<cdh <major.minor version>/

    For example: http://xyznode46.unraveldata.com:3000/parcels/cdh6.3/

    • <unravel-host> is the hostname or LAN IP address of Unravel. In a multi-cluster scenario, this would be the host where the log_receiver daemon is running.

    • <port> is the Unravel UI port. The default is 3000. In case you have customized the default port, you can add that port number.

    • <cdh-version> is your version of CDH. For example, cdh5.16 or cdh6.3.

      You can go to http://<unravel-host>:<port>/parcels/ directory (For example: http://xyznode46.unraveldata.com:3000/parcels) and copy the exact directory name of your CDH version (CDH<major.minor>).

    Note

    If you're using Active Directory Kerberos, unravel-host must be a fully qualified domain name or IP address.

    Tip

    If you are running more than one version of CDH (for example, you have multiple clusters), you can add more than one parcel entry for unravel-host.

  4. Click Save Changes.

  5. In the Cloudera Manager, click Check for new parcels, find the UNRAVEL_SENSOR parcel that you want to distribute, and click the corresponding Download button.

  6. In the Cloudera Manager, from Location > Parcel Name, find the UNRAVEL_SENSOR parcel that you want to distribute and click the corresponding Download button.

  7. After the parcel is downloaded, click the corresponding Distribute button. This will distribute the parcel to all the hosts.

  8. After the parcel is distributed, click the corresponding Activate button. The status column will now display Distributed, Activated.

    Note

    If you have an old sensor parcel from Unravel, you must deactivate it now.

2. Put the Hive Hook JAR in AUX_CLASSPATH

  1. In Cloudera Manager, select the target cluster from the drop-down, click Hive >Configuration, and search for hive-env.

  2. In Gateway Client Environment Advanced Configuration Snippet (Safety Valve) for hive-env.sh, click View as text and enter the following exactly as shown, with no substitutions:

    AUX_CLASSPATH=${AUX_CLASSPATH}:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar

  3. If Sentry is enabled, grant privileges on the JAR files to the Sentry roles that run Hive queries.

    Sentry commands may also be needed to enable access to the Hive Hook JAR file. Grant privileges on the JAR files to the roles that run hive queries. Log in to Beeline as user hive and use the Hive SQL GRANT statement to do so.

    For example (substitute role as appropriate),

    GRANT ALL ON URI 'file:///opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar' TO ROLE <role>
3. Oozie: Copy Hive Hook and BTrace JARs to HDFS shared library path

  1. In Cloudera Manager, select the target cluster from the drop-down, click Oozie >Configuration and check the path shown in ShareLib Root Directory.

  2. From a terminal application on the Unravel node (edge node in case of multi-cluster.), pick up the ShareLib Root Directory directory path with the latest timestamp.

    hdfs dfs -ls <path to ShareLib directory>

    For example: hdfs dfs -ls /user/oozie/share/lib/

    Important

    The jars must be copied to the lib directory (with the latest timestamp), which is shown in ShareLib Root Directory.

  3. Copy the Hive Hook JAR /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar and the Btrace JAR, /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar to the specified path in ShareLib Root Directory.

    For example, if the path specified in ShareLib Root Directory. is /user/oozie, run the following commands to copy the JAR files.

    hdfs dfs -copyFromLocal /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar /user/oozie/share/lib/<latest timestamp lib directory>/

    For example: hdfs dfs -copyFromLocal /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar /user/oozie/share/lib/lib_20210326035616/

    hdfs dfs -copyFromLocal /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar /user/oozie/share/lib/<latest timestamp lib directory>/

    For example: hdfs dfs -copyFromLocal /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar /user/oozie/share/lib/lib_20210326035616/

  4. From a terminal application, copy the Hive Hook JAR /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/unravel_hive_hook.jar and the Btrace JAR, /opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar to the specified path in ShareLib Root Directory.

    Caution

    Jobs controlled by Oozie 2.3+ fail if you do not copy the Hive Hook and BTrace JARs to the HDFS shared library path.

4. Set Hive Hook configuration

  1. On the Cloudera Manager, click Hive service and then click the Configuration tab.

  2. Search for hive-site.xml, which will lead to the Hive Client Advanced Configuration Snippet for hive-site.xml section.

  3. Specify the hive hook configurations. You have the option to either use the XML text field or Editor to specify the hive hook configuration.

    • Option 1: XML text field

      Click View as XML to open the XML text field and copy-paste the following.

      <property>
  <name>com.unraveldata.host</name>
  <value><UNRAVEL HOST NAME></value> 
  <description>Unravel hive-hook processing host</description>
</property>
<property>
  <name>com.unraveldata.hive.hook.tcp</name>
  <value>true</value>
</property>
<property>
  <name>com.unraveldata.hive.hdfs.dir</name>
  <value>/user/unravel/HOOK_RESULT_DIR</value>
  <description>destination for hive-hook, Unravel log processing</description>
</property>
<property>
  <name>hive.exec.driver.run.hooks</name>
<value>com.unraveldata.dataflow.hive.hook.UnravelHiveHook</value>
  <description>for Unravel, from unraveldata.com</description>
</property>
<property>
  <name>hive.exec.pre.hooks</name>  <value>com.unraveldata.dataflow.hive.hook.UnravelHiveHook</value>
  <description>for Unravel, from unraveldata.com</description>
</property>
<property>
  <name>hive.exec.post.hooks</name>  <value>com.unraveldata.dataflow.hive.hook.UnravelHiveHook</value>
  <description>for Unravel, from unraveldata.com</description>
</property>
<property>
  <name>hive.exec.failure.hooks</name>  <value>com.unraveldata.dataflow.hive.hook.UnravelHiveHook</value>
  <description>for Unravel, from unraveldata.com</description>
</property>

      Ensure to replace UNRAVEL HOST NAME with the Unravel hostname. Replace The Unravel Host Name with the hostname of the edge node in case of a multi-cluster deployment.

    • Option 2: Editor:

      Click + and enter the property, value, and description (optional).

      Property

      Value

      Description

      com.unraveldata.host

      Replace with Unravel hostname or with the hostname of the edge node in case of a multi-cluster deployment.

      Unravel hive-hook processing host

      com.unraveldata.hive.hook.tcp

      true

      Hive hook tcp protocol.

      com.unraveldata.hive.hdfs.dir

      /user/unravel/HOOK_RESULT_DIR

      Destination directory for hive-hook, Unravel log processing.

      hive.exec.driver.run.hooks

      com.unraveldata.dataflow.hive.hook.UnravelHiveHook

      Hive hook

      hive.exec.pre.hooks

      com.unraveldata.dataflow.hive.hook.UnravelHiveHook

      Hive hook

      hive.exec.post.hooks

      com.unraveldata.dataflow.hive.hook.UnravelHiveHook

      Hive hook

      hive.exec.failure.hooks

      com.unraveldata.dataflow.hive.hook.UnravelHiveHook

      Hive hook

    Note

    If you configure CDH with Cloudera Navigator's safety valve setting, you must edit the following keys and append the value com.unraveldata.dataflow.hive.hook.UnravelHiveHook without any space.

    • hive.exec.post.hooks

    • hive.exec.pre.hooks

    • hive.exec.failure.hooks

    For example:

    <property>  
<name>hive.exec.post.hooks</name>  
<value>com.unraveldata.dataflow.hive.hook.UnravelHiveHook,com.cloudera.navigator.audit.hive.HiveExecHookContext,org.apache.hadoop.hive.ql.hooks.LineageLogger</value>  
<description>for Unravel, from unraveldata.com</description>
</property>

  4. Similarly, ensure to add the same hive hook configurations in HiveServer2 Advanced Configuration Snippet for hive-site.xml.

  5. Optionally, add a comment in Reason for change and then click Save Changes.

  6. From the Cloudera Manager page, click the Stale configurations icon (DeployGlyph.png) to deploy the configuration and restart the Hive services.

  7. Check Unravel UI to see if all Hive queries are running.

    • If queries are running fine and appearing in Unravel UI, then you have successfully added the hive hooks configurations.

    • If queries are failing with a class not found error or permission problems:

      • Undo the hive-site.xml changes in Cloudera Manager.

      • Deploy the hive client configuration.

      • Restart the Hive service.

      • Follow the steps in Troubleshooting.

5. Configure Spark properties in spark-defaults.conf

  1. In Cloudera Manager, select the target cluster and then click Spark.

  2. Select Configuration.

  3. Search for spark-defaults.

  4. In Spark Client Advanced Configuration Snippet (Safety Valve) for spark-conf/spark-defaults.conf, enter the following text, replacing placeholders with your particular values:

    spark.unravel.server.hostport=unravel-host:<port>
spark.driver.extraJavaOptions=-javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=config=driver,libs=spark-version
spark.executor.extraJavaOptions=-javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=config=executor,libs=spark-version
spark.eventLog.enabled=true

    Note

    In case you are using JDK version 9 or later, ensure to add the following to the existing JAVA options:

    --add-exports java.base/jdk.internal.perf=ALL-UNNAMED 
--add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED
--add-exports java.management/sun.management.counter.perf=ALL-UNNAMED
--add-exports java.management/sun.management.counter=ALL-UNNAMED

    For example, the complete JAVA options are specified as follows:

    -javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=config=executor,libs=spark-version --add-exports java.base/jdk.internal.perf=ALL-UNNAMED  --add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED --add-exports java.management/sun.management.counter.perf=ALL-UNNAMED --add-exports java.management/sun.management.counter=ALL-UNNAMED

    • <unravel-host>: Specify the Unravel hostname. In the case of multi-cluster deployment use the FQDN or logical hostname of the edge node for unravel-host.

    • <Port>: 4043 is the default port. If you have customized the ports, you can specify that port number here.

    • <spark-version>: For spark-version, use a Spark version that is compatible with this version of Unravel. You can check the Spark version with the spark-submit --version command and specify the same version.

  5. Click Save changes.

  6. Click (DeployGlyph.png) or use the Actions pull-down menu to deploy the client configuration. Your spark-shell will ensure new JVM containers are created with the necessary extraJavaOptions for the Spark drivers and executors.

  7. Enable Spark streaming.

  8. Check Unravel UI to see if all Spark jobs are running.

    • If jobs are running and appearing in Unravel UI, you have deployed the Spark jar successfully.

    • If queries are failing with a class not found error or permission problems:

      • Undo the spark-defaults.conf changes in Cloudera Manager.

      • Deploy the client configuration.

      • Investigate and fix the issue.

      • Follow the steps in Troubleshooting.

Note

If you have YARN-client mode applications, the default Spark configuration is not sufficient, because the driver JVM starts before the configuration set through the SparkConf is applied. For more information, see Apache Spark Configuration. In this case, configure the Unravel Sensor for Spark to profile specific Spark applications only (in other words, per-application profiling rather than cluster-wide profiling).

6. Optional: Configure YARN-MapReduce JVM sensor cluster-wide

  1. In Cloudera Manager, go to YARN service.

  2. Select the Configuration tab.

  3. Search for Application Master Java Opts Base and concatenate the following XML block properties snippet (ensure to start with a space and add below).

    Note

    Make sure that "-" is a minus sign. You need to modify the value of unravel-host with your Unravel Server IP address or a fully qualified DNS. For multi-host Unravel installation, use the IP address of Host2.

    -javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=libs=mr -Dunravel.server.hostport=unravel-host:4043

    Note

    In case you are using JDK version 9 or later, ensure to add the following to the existing JAVA options:

    --add-exports java.base/jdk.internal.perf=ALL-UNNAMED 
--add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED
--add-exports java.management/sun.management.counter.perf=ALL-UNNAMED
--add-exports java.management/sun.management.counter=ALL-UNNAMED

    For example, the complete JAVA options are specified as follows:

    -javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=config=executor,libs=spark-version --add-exports java.base/jdk.internal.perf=ALL-UNNAMED  --add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED --add-exports java.management/sun.management.counter.perf=ALL-UNNAMED --add-exports java.management/sun.management.counter=ALL-UNNAMED

  4. Search for MapReduce Client Advanced Configuration Snippet (Safety Valve) for mapred-site.xml in the middle of the page.

  5. Enter the following XML four-block properties snippet to Gateway Default Group. (Click View as XML.)

    <property>
<name>mapreduce.task.profile</name>
<value>true</value>
</property> 
<property>
<name>mapreduce.task.profile.maps</name>
<value>0-5</value>
</property> 
<property>
<name>mapreduce.task.profile.reduces</name>
<value>0-5</value>
</property> 
<property>
<name>mapreduce.task.profile.params</name>
<value>-javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=libs=mr -Dunravel.server.hostport=<unravel-host>:4043</value></property>

    Note

    Replace unravel-host to Unravel hostname. In case of multi-cluster, add the hostname of the edge node.

    Note

    In case you are using JDK version 9 or later, ensure to add the following to the existing JAVA options:

    --add-exports java.base/jdk.internal.perf=ALL-UNNAMED 
--add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED
--add-exports java.management/sun.management.counter.perf=ALL-UNNAMED
--add-exports java.management/sun.management.counter=ALL-UNNAMED

    For example, the complete JAVA options are specified as follows:

    -javaagent:/opt/cloudera/parcels/UNRAVEL_SENSOR/lib/java/btrace-agent.jar=config=executor,libs=spark-version --add-exports java.base/jdk.internal.perf=ALL-UNNAMED  --add-opens jdk.management/com.sun.management.internal=ALL-UNNAMED --add-exports java.management/sun.management.counter.perf=ALL-UNNAMED --add-exports java.management/sun.management.counter=ALL-UNNAMED

  6. Save the changes.

  7. Deploy the client configuration by using the Actions pull-down menu.

    Cloudera Manager will specify a restart which is not necessary to effect these changes. (Click Restart Stale Services if that is visible. However, you can also perform this later when you have planned maintenance.)

Tip

The restart is important for the MR sensor to be picked up by queries submitted via Hiveserver2.

Use the Unravel UI to monitor the situation. When you view the MapReduce APM page for any completed MRjob you should see mappers and reducers in the Resource Usage tab.

7. Retrieve Impala data from Cloudera Manager

Impala properties are automatically configured. Refer to Impala properties for the list of properties that are automatically configured. If it is not set already by auto-configuration, set the properties as follows:

<Unravel installation directory>/manager config properties set <PROPERTY> <VALUE>

For example,

<Unravel installation directory>/manager config properties set com.unraveldata.data.source cm 
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.url http://my-cm-url  
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.username mycmname 
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.password mycmpassword

For multi-cluster, use the following format and set these on the edge node:

<Unravel installation directory>/manager config properties set com.unraveldata.data.source cm 
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.url http://my-cm-url  
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.username mycmname 
<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.password mycmpassword

Note

By default, the Impala sensor task is enabled. To disable it, you can edit the following property as follows:

<Unravel installation directory>/manager config properties set com.unraveldata.sensor.tasks.disabled iw

Optionally, you can change the Impala lookback window. By default, when Unravel Server starts, it retrieves the last 5 minutes of Impala queries. To change this, do the following:

Change the value for com.unraveldata.cloudera.manager.impala.look.back.minutes property.

<Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.impala.look.back.minutes -<period>

For example: <Unravel installation directory>/manager config properties set com.unraveldata.cloudera.manager.impala.look.back.minutes -7

Note

Include a minus sign in front of the new value.

8. Enable Impala Monitoring

Refer to Monitoring Impala.

9. Add more configurations

Set additional Unravel configurations.

Creating an alternate principal

For quick initial installation, you can use the hdfs principal and its keytab. However, for production use you may want to create an alternate principal that has restricted access to specific areas and use its corresponding keytab. This topic explains how to do this.

You can name the alternate principal whatever you prefer; these steps name it unravel. Its name doesn't need to be the same as the local username.

The steps apply only to CDH and have been tested using Cloudera Manager with the recommended Sentry configuration.

  1. Check the HDFS default umask.

    For access via ACL, the group part of the HDFS default umask needs to have read and execute access. This allows Unravel to see subdirectories and read files. The default umask setting on HDFS for both CDH and HDP is 022. The middle digit controls the group mask, and ACLs are masked using this default group mode.

    You can check the HDFS umask setting from either Cloudera Manager or in hdfs-site.xml:

    • In Cloudera Manager, check the value of dfs.umaskmode and make sure the middle digit is 2 or 0.

    • In hdfs-site.xml file search for fs.permissions.umask-mode and make sure the middle digit is 2 or 0.

  2. Enable ACL inheritance.

    In Cloudera Manager's HDFS configuration, search for namenode advanced configuration snippet, and set its dfs.namenode.posix.acl.inheritance.enabled property to true in hdfs-site.xml. This is a workaround for an issue where HDFS was not compliant with the Posix standard for ACL inheritance. For details, see Apache JIRA HDFS-6962. Cloudera backported the fix for this issue into CDH5.8.4, CDH5.9.1, and later, setting dfs.namenode.posix.acl.inheritance.enabled to false in Hadoop 2.x and true in Hadoop 3.x.

  3. Restart the cluster to effect the change of dfs.namenode.posix.acl.inheritance.enabled to true.

  4. Change the ACLs of the target HDFS directories.

    Run the following commands as global hdfs to change the ACLs of the following HDFS directories. Run these in the order presented.

    1. Set the ACL for future directories.

      Note

      Be sure to set the permissions at the /user/history level. Files are first written to an intermediate_done folder under /user/history and then moved to /user/history/done.

      hadoop fs -setfacl -R -m default:user:unravel:r-x /user/spark/applicationHistory
hadoop fs -setfacl -R -m default:user:unravel:r-x /user/history
hadoop fs -setfacl -R -m default:user:unravel:r-x /tmp/logs
hadoop fs -setfacl -R -m default:user:unravel:r-x /user/hive/warehouse

      If you have Spark2 installed, set the ACL of the Spark2 application history folder:

      hadoop fs -setfacl -R -m default:user:unravel:r-x /user/spark/spark2ApplicationHistory

    2. Set ACL for existing directories.

      hadoop fs -setfacl -R -m user:unravel:r-x /user/spark/applicationHistory
hadoop fs -setfacl -R -m user:unravel:r-x /user/history
hadoop fs -setfacl -R -m user:unravel:r-x /tmp/logs
hadoop fs -setfacl -R -m user:unravel:r-x /user/hive/warehouse

      If you have Spark2 installed, set the ACL of the Spark2 application history folder:

      hadoop fs -setfacl -R -m user:unravel:r-x /user/spark/spark2ApplicationHistory

  5. Verify the ACL of the target HDFS directories.

    hdfs dfs -getfacl /user/spark/applicationHistory
hdfs dfs -getfacl /user/spark/spark2ApplicationHistory
hdfs dfs -getfacl /user/history
hdfs dfs -getfacl /tmp/logs
hdfs dfs -getfacl /user/hive/warehouse

  6. On the Unravel Server, verify HDFS permission on folders as the target user (unravel, hdfs, mapr, or custom) with a valid kerberos ticket corresponding to the keytab principal.

    sudo -u unravel kdestroy
sudo -u unravel kinit -kt keytab-file principal
sudo -u unravel hadoop fs -ls /user/history
sudo -u unravel hadoop fs -ls /tmp/logs
sudo -u unravel hadoop fs -ls /user/hive/warehouse

  7. Find and verify the keytab:

    klist -kt keytab-file

    Warning

    If you're using KMS and HDFS encryption and the hdfs principal, you might need to adjust kms-acls.xml permissions in Cloudera Manager for DECRYPT_EEK if access is denied. In particular, the "done" directory might not allow decryption of logs by the hdfs principal.

    If you're using "JNI" based groups for HDFS (a setting in Cloudera Manager), you need to add this line to /usr/local/unravel/etc/unravel.ext.sh:

    export LD_LIBRARY_PATH=/opt/cloudera/parcels/CDH/lib/hadoop/lib/native

  8. If Kerberos is enabled, set the new values for keytab-file and principal:

    <Unravel installation directory>/manager config kerberos set --keytab /etc/security/keytabs/unravel.service.keytab --principal unravel/server@example.com

<Unravel installation directory>/manager config kerberos enable

    Important

    Whenever you change Kerberos tokens or principal, restart all services, <installation directory>/manager restart.

References

For more information on creating permanent functions, see Cloudera documentation.

In this section: