Apache Ambari Project WebsiteAmbari User Guide

 

Apache Ambari

Installing Hadoop Using Apache Ambari

2013-07-16

Contents

1. Getting Ready to Install
1. Understand the Basics
2. Meet Minimum System Requirements
2.1. Hardware Recommendations
2.2. Operating Systems Requirements
2.3. Browser Requirements
2.4. Software Requirements
2.5. Database Requirements
3. Decide on Deployment Type
4. Collect Information
5. Prepare the Environment
5.1. Check Existing Installs
5.2. Set Up Password-less SSH
5.3. Enable NTP on the Cluster and on the Browser Host
5.4. Check DNS
5.5. Disable SELinux
5.6. Disable iptables
5.7. Disable PackageKit
5.8. Check umask Value
6. Optional: Configure the Local Repositories
2. Running the Installer
1. Set Up the Bits
1.1. RHEL/CentOS 5.x
1.2. RHEL/CentOS 6.x
1.3. SLES 11
2. Set Up the Server
2.1. Setup Options
3. Optional: Set Up LDAP or Active Directory Authentication
4. Optional: Change the Ambari Server Port
5. Start the Ambari Server
3. Installing, Configuring, and Deploying the Cluster
1. Log into Apache Ambari
2. Welcome
3. Select Stack
4. Install Options
5. Confirm Hosts
6. Choose Services
7. Assign Masters
8. Assign Slaves and Clients
9. Customize Services
9.1. Properties That Depend on Service Usernames/Groups
9.2. Recommended Memory Configurations for the MapReduce Service
10. Review
11. Install, Start and Test
12. Summary
4. Troubleshooting Ambari Deployments
1. Getting the Logs
2. Quick Checks
3. Specific Issues
3.1. Problem: Browser crashed before Install Wizard completed
3.2. Problem: Install Wizard reports that the cluster install has failed
3.3. Problem: “Unable to create new native thread” exceptions in HDFS DataNode logs or those of any system daemon
3.4. Problem: The “yum install ambari-server” Command Fails
3.5. Problem: HDFS Smoke Test Fails
3.6. Problem: The HCatalog Daemon Metastore Smoke Test Fails
3.7. Problem: MySQL and Nagios fail to install on RightScale CentOS 5 images on EC2
3.8. Problem: Trouble starting Ambari on system reboot
3.9. Problem: Metrics and Host information display incorrectly in Ambari Web
3.10. Problem: On SUSE 11 Ambari Agent crashes within the first 24 hours
3.11. Problem: Attempting to Start HBase REST server causes either REST server or Ambari Web to fail
3.12. Problem: Multiple Ambari Agent processes are running, causing re-register
3.13. Problem: Some graphs do not show a complete hour of data until the cluster has been running for an hour
3.14. Problem: After performing a cluster install the Nagios server is not started
3.15. Problem: A service with a customized service user is not appearing properly in Ambari Web
3.16. Problem: Updated configuration changes are not pushed to client/gateway nodes
3.17. Problem: Trying to upgrade Ambari Server on SLES produces error
5. Appendix: Installing Ambari Agents Manually
1. RHEL/CentOS v. 5.x and 6.x
2. SLES
6. Appendix: Using Custom Hostnames
7. Appendix: Upgrading Operating Systems on an Ambari-based Hadoop Installation
8. Appendix: Upgrading Ambari Server from 1.2.0/1.2.1/1.2.2/1.2.2.5/1.2.3.6/1.2.3.7 to 1.2.4
9. Appendix: Upgrading the HDP Stack from 1.2.0/1.2.1 to 1.3.0
1. Preparing for the Upgrade
2. Setting Up the Ambari Repository
3. Upgrading Ambari Server and Agents
4. Upgrading the Stack
10. Appendix: Configuring Ports
1. HDFS Ports
2. MapReduce Ports
3. Hive Ports
4. HBase Ports
5. WebHCat Port
6. Ganglia Ports
7. MySQL Port
11. Appendix: Moving the Ambari Server
1. Back up Current Data
2. Update Agents
3. Install New Server and Populate the Databases
12. Appendix: Using Non-Default Databases
1. Hive/HCatalog
1.1. Troubleshooting Hive/HCatalog
2. Oozie
2.1. Troubleshooting Oozie
3. Ambari
3.1. Troubleshooting Ambari

List of Tables

2.1. Download the repo
2.2. Ambari tarball
2.3. Ambari Server Properties
3.1. HDFS Settings: Advanced
3.2. MapReduce Settings: Advanced
6.1. ambari-agent.ini
9.1. Ambari tarball
10.1. HDFS Ports
10.2. MapReduce Ports
10.3. Hive Ports
10.4. HBase Ports
10.5. WebHCat Port
10.6. Ganglia Ports
10.7. MySQL Port
Index
loading table of contents...