Hue Tarball Installation Guide ============================== Introduction ------------ Hue is a graphical user interface to operate and develop applications for Apache Hadoop. Hue applications are collected into a desktop-style environment and delivered as a Web application, requiring no additional installation for individual users. This guide describes how to install and configure a Hue tarball. For information about installing Hue packages, see https://wiki.cloudera.com/display/DOC/Hue+Installation[Installing Hue]. There is also a companion SDK guide that describes how to develop new Hue applications: http://archive.cloudera.com/cdh3/hue/sdk/sdk.html[Hue SDK Documentation] IMPORTANT: Hue requires the Hadoop contained in Cloudera's Distribution including Apache Hadoop (CDH), version 3 update 4 or later. .Conventions Used in this Guide: * Commands that must be run with +root+ permission have a +#+ command prompt. * Commands that do not require +root+ permission have a +$+ command prompt. Hue Installation Instructions ----------------------------- The following instructions describe how to install the Hue tarball on a multi-node cluster. You need to also install CDH and update some Hadoop configuration files before running Hue. Install Hue ~~~~~~~~~~~ Hue consists of a web service that runs on a special node in your cluster. Choose one node where you want to run Hue. This guide refers to that node as the _Hue Server_. For optimal performance, this should be one of the nodes within your cluster, though it can be a remote node as long as there are no overly restrictive firewalls. For small clusters of less than 10 nodes, you can use your existing master node as the Hue Server. You can download the Hue tarball here: http://github.com/cloudera/hue/downloads/ Hue Dependencies ^^^^^^^^^^^^^^^^ Hue employs some Python modules which use native code and requires certain development libraries be installed on your system. To install from the tarball, you must have the following installed: .Required Dependencies [grid="rows"] ``~ Redhat,Debian ~~~~~~~~~~ gcc,gcc libxml2-devel,libxml2-dev libxslt-devel,libxslt-dev cyrus-sasl-devel,libsasl2-dev mysql-devel,libmysqlclient-dev python-devel,python-dev python-setuptools,python-setuptools python-simplejson,python-simplejson sqlite-devel,libsqlite3-dev ant,ant ~~~~~~~~~~ Build ^^^^^ Configure `$PREFIX` with the path where you want to install Hue by running: $ PREFIX=/path/to/install/into make install You can install Hue anywhere on your system - it does not need root permission although additional third-party SDK applications may. It is a good practice to create a new user for Hue and either install Hue in that user's home directory, or in a directory within `/usr/share`. Troubleshooting the Hue Tarball Installation ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ .Q: I moved my Hue installation from one directory to another and now Hue no longer functions correctly. A: Due to the use of absolute paths by some Python packages, you must run a series of commands if you move your Hue installation. In the new location, run: ---- $ rm app.reg $ rm -r build $ make apps ---- .Q: Why does "make install" compile other pieces of software? A: In order to ensure that Hue is stable on a variety of distributions and architectures, it installs a Python virtual environment which includes its dependencies. This ensures that the software can depend on specific versions of various Python libraries and you don't have to be concerned about missing software components. Install Hadoop from CDH ~~~~~~~~~~~~~~~~~~~~~~~ To use Hue, you must install and run Hadoop from CDH3u4 or later. If you are not running this version of CDH or later, upgrade your cluster before proceeding. .Dependency on CDH Components [grid="rows"] `-------------.----------.----------------------------------.------------------------ Component Required Applications Notes ------------------------------------------------------------------------------------- HDFS Yes Core, Filebrowser HDFS access through WebHdfs or HttpFS MR1 No JobBrowser, JobDesigner*, Beeswax* Job information access through hue-plugins Yarn No JobDesigner*, Beeswax* Transitive dependency via Hive or Oozie Oozie No JobDesigner Oozie access through REST API Hive No Beeswax Beeswax uses the Hive client libraries Flume No Shell Optionally provides access to the Flume shell HBase No Shell Optionally provides access to the HBase shell Pig No Shell Optionally provides access to the Pig shell ------------------------------------------------------------------------------------- [*] Transitive dependency Hadoop Configuration ~~~~~~~~~~~~~~~~~~~~ Configure WebHdfs ^^^^^^^^^^^^^^^^^ You need to enable WebHdfs or run an HttpFS server. To turn on WebHDFS, add this to your `hdfs-site.xml` and *restart* your HDFS cluster. Depending on your setup, your `hdfs-site.xml` might be in `/etc/hadoop/conf`. dfs.webhdfs.enabled true If you place your Hue Server outside the Hadoop cluster, you can run an HttpFS server to provide Hue access to HDFS. The HttpFS service requires only one port to be opened to the cluster. Configure MapReduce 0.20 (MR1) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Hue communicates with the JobTracker via the Hue plugins, which is a jar file that you place in your MapReduce `lib` directory. If you JobTracker and Hue are located on the same host, copy it over. If you are using CDH3, your MapReduce library directory might be in `/usr/lib/hadoop/lib`. $ cd /usr/share/hue $ cp desktop/libs/hadoop/java-lib/hue-plugins-*.jar /usr/lib/hadoop-0.20-mapreduce/lib If you JobTracker runs on a different host, you need to `scp` the Hue plugins jar to the JobTracker host. Then add this to your `mapred-site.xml` and *restart* your JobTracker. Depending on your setup, your `hdfs-site.xml` might be in `/etc/hadoop/conf`. jobtracker.thrift.address 0.0.0.0:9290 mapred.jobtracker.plugins org.apache.hadoop.thriftfs.ThriftJobTrackerPlugin Comma-separated list of jobtracker plug-ins to be activated. You can confirm that the plugins are running correctly by tailing the daemon logs: $ tail --lines=500 /var/log/hadoop-0.20/hadoop*jobtracker*.log | grep ThriftPlugin 2009-09-28 16:30:44,337 INFO org.apache.hadoop.thriftfs.ThriftPluginServer: Starting Thrift server 2009-09-28 16:30:44,419 INFO org.apache.hadoop.thriftfs.ThriftPluginServer: Thrift server listening on 0.0.0.0:9290 Further Hadoop Configuration and Caveats ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ `HADOOP_CLASSPATH` Caveat ^^^^^^^^^^^^^^^^^^^^^^^^^ If you are setting `$HADOOP_CLASSPATH` in your `hadoop-env.sh`, be sure to set it in such a way that user-specified options are preserved. For example: Correct: # HADOOP_CLASSPATH=:$HADOOP_CLASSPATH Incorrect: # HADOOP_CLASSPATH= This enables certain components of Hue to add to Hadoop's classpath using the environment variable. `hadoop.tmp.dir` ^^^^^^^^^^^^^^^^ If your users are likely to be submitting jobs both using Hue and from the same machine via the command line interface, they will be doing so as the `hue` user if they're using Hue and via their own user account on the command line. This leads to some contention on the directory specified by `hadoop.tmp.dir`, which defaults to `/tmp/hadoop-${user.name}`. Specifically, `hadoop.tmp.dir` is used to unpack jars in `bin/hadoop jar`. One work around to this is to set `hadoop.tmp.dir` to `/tmp/hadoop-${user.name}-${hue.suffix}` in the core-site.xml file: hadoop.tmp.dir /tmp/hadoop-${user.name}${hue.suffix} Unfortunately, when the variable is unset, you'll end up with directories named `/tmp/hadoop-user_name-${hue.suffix}` in `/tmp`. Despite that, Hue will still work. IMPORTANT: The Beeswax server writes into a local directory on the Hue machine that is specified by `hadoop.tmp.dir` to unpack its jars. That directory needs to be writable by the `hue` user, which is the default user who starts Beeswax Server, or else Beeswax server will not start. You may also make that directory world-writable. Configuring Your Firewall for Hue ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Hue currently requires that the machines within your cluster can connect to each other freely over TCP. The machines outside your cluster must be able to open TCP port 8888 on the Hue Server (or the configured Hue web HTTP port) to interact with the system. Hive Configuration ~~~~~~~~~~~~~~~~~~ Hue's Beeswax application helps you use Hive to query your data. It depends on a Hive installation on your system. Please read this section to ensure a proper integration. Your Hive data is stored in HDFS, normally under `/user/hive/warehouse` (or any path you specify as `hive.metastore.warehouse.dir` in your `hive-site.xml`). Make sure this location exists and is writable by the users whom you expect to be creating tables. `/tmp` (on the local file system) must be world-writable, as Hive makes extensive use of it. [NOTE] If you used the embedded Hive MetaStore functionality of Beeswax in Hue from versions prior to Hue 1.2, read this section. Hue 1.2 includes changes in the Hive MetaStore schema that are part of the Hive 0.7 release. If you want to use Beeswax in Hue 1.2, it is imperative that you upgrade the Hive MetaStore schema by running the appropriate schema upgrade script located in the `apps/beeswax/hive/scripts/metastore/upgrade` directory in the Hue installation. Scripts for Derby and MySQL databases are available. If you are using a different database for your MetaStore, you will need to provide your own upgrade script. No Existing Hive Installation ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Familiarize yourself with the configuration options in `hive-site.xml`. See https://ccp.cloudera.com/display/CDHDOC/Hive+Installation#HiveInstallation-ConfiguringtheHiveMetastore[Hive Installation and Configuration]. Having a `hive-site.xml` is optional but often useful, particularly on setting up a http://wiki.apache.org/hadoop/Hive/AdminManual/MetastoreAdmin[metastore]. You may store the `hive-site.xml` in `/etc/hue/conf`, or instruct Beeswax to locate it using the `hive_conf_dir` configuration variable. See `/etc/hue/conf/hue.ini`. Existing Hive Installation ^^^^^^^^^^^^^^^^^^^^^^^^^^ In `/etc/hue/conf/hue.ini`, modify `hive_conf_dir` to point to the directory containing `hive-site.xml`. Configuring Hue --------------- Hue ships with a default configuration that will work for pseudo-distributed clusters. If you are running on a real cluster, you must make a few changes to the `/etc/hue/hue.ini` configuration file. The following sections describe the key configuration options you must make to configure Hue. [TIP] .Listing all Configuration Options ============================================================ To list all available configuration options, run: $ /usr/share/hue/build/env/bin/hue config_help | less This commands outlines the various sections and options in the configuration, and provides help and information on the default values. ============================================================ [TIP] .Viewing Current Configuration Options ============================================================ To view the current configuration from within Hue, open: http:///dump_config ============================================================ [TIP] .Using Multiple Files to Store Your Configuration ============================================================ Hue loads and merges all of the files with extension `.ini` located in the `/etc/hue` directory. Files that are alphabetically later take precedence. ============================================================ Web Server Configuration ~~~~~~~~~~~~~~~~~~~~~~~~ These configuration variables are under the `[desktop]` section in the `/etc/hue/hue.ini` configuration file. Specifying the Hue HTTP Address ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ Hue uses a Spawning or a CherryPy web server (configurable). You can use the following options to change the IP address and port that the web server listens on. The default setting is port 8888 on all configured IP addresses. # Webserver listens on this address and port http_host=0.0.0.0 http_port=8888 Specifying the Secret Key ^^^^^^^^^^^^^^^^^^^^^^^^^ For security, you should also specify the secret key that is used for secure hashing in the session store. Enter a long series of random characters (30 to 60 characters is recommended). secret_key=jFE93j;2[290-eiw.KEiwN2s3['d;/.q[eIW^y#e=+Iei*@Mn host.key ### Create a self-signed certificate $ openssl req -new -x509 -nodes -sha1 -key host.key > host.cert [NOTE] .Self-signed Certificates and File Uploads ============================================================ To upload files using the Hue File Browser over HTTPS requires using a proper SSL Certificate. Self-signed certificates don't work. ============================================================ Hue Configuration for Hadoop ~~~~~~~~~~~~~~~~~~~~~~~~~~~~ These configuration variables are under the `[hadoop]` section in the `/etc/hue/hue.ini` configuration file. HDFS Cluster ^^^^^^^^^^^^ Hue only support one HDFS cluster currently. That cluster should be defined under the `[[[default]]]` sub-section. fs_defaultfs:: This is the equivalence of `fs.defaultFS` (aka `fs.default.name`) in Hadoop configuration. webhdfs_url:: You can also set this to be the HttpFS url. The default value is the HTTP port on the NameNode. hadoop_hdfs_home:: This is the home of your Hadoop HDFS installation. It is the root of the Hadoop untarred directory, or usually `/usr/lib/hadoop`. hadoop_bin:: Use this as the HDFS Hadoop launcher script, which is usually `/usr/bin/hadoop`. hadoop_conf_dir:: This is the configuration directory of the HDFS, typically `/etc/hadoop/conf`. MapReduce (MR1) Cluster ^^^^^^^^^^^^^^^^^^^^^^^ Hue only support one MapReduce cluster currently. That cluster should be defined under the `[[[default]]]` sub-section. Note that JobBrowser only works with MR1. jobtracker_host:: The host running the JobTracker. jobtracker_port:: The port for the JobTracker IPC service. submit_to:: If your Oozie is configured with to talk to a 0.20 MapReduce service, then set this to `true`. Hue will be submitting jobs to this MapReduce cluster. hadoop_mapred_home:: This is the home of your Hadoop MapReduce installation. It is the root of the Hadoop MR1 untarred directory, or the root of the Hadoop 0.20 untarred directory, or `/usr/lib/hadoop-0.20-mapreduce` for CDH packages. If `submit_to` is true for this cluster, this config value becomes the `$HADOOP_MAPRED_HOME` for BeeswaxServer and child shell processes. hadoop_bin:: Use this as the MR1 Hadoop launcher script, which is usually `/usr/bin/hadoop`. hadoop_conf_dir:: This is the configuration directory of the MR1 service, typically `/etc/hadoop/conf`. Yarn (MR2) Cluster ^^^^^^^^^^^^^^^^^^ Hue only support one Yarn cluster currently. That cluster should be defined under the `[[[default]]]` sub-section. resourcemanager_host:: The host running the ResourceManager. resourcemanager_port:: The port for the ResourceManager IPC service. submit_to:: If your Oozie is configured with to talk to a Yarn cluster, then set this to `true`. Hue will be submitting jobs to this Yarn cluster. But note that JobBrowser will not be able to show MR2 jobs. hadoop_mapred_home:: This is the home of your Hadoop MapReduce installation. It is the root of the Hadoop 0.23 untarred directory, or `/usr/lib/hadoop-mapreduce` for CDH packages. If `submit_to` is true for this cluster, this config value becomes the `$HADOOP_MAPRED_HOME` for BeeswaxServer and child shell processes. hadoop_bin:: Use this as the Yarn/MR2 Hadoop launcher script, which is usually `/usr/bin/hadoop`. hadoop_conf_dir:: This is the configuration directory of the Yarn/MR2 service, typically `/etc/hadoop/conf`. Beeswax Configuration ~~~~~~~~~~~~~~~~~~~~~ In the `[beeswax]` section of the configuration file, you can _optionally_ specify the following: beeswax_server_host:: The hostname or IP that the Beeswax Server should bind to. By default it binds to `localhost`, and therefore only serves local IPC clients. hive_home_dir:: The base directory of your Hive installation. hive_conf_dir:: The directory containing your `hive-site.xml` Hive configuration file. beeswax_server_heapsize:: The heap size (-Xmx) of the Beeswax Server. JobDesigner Configuration ~~~~~~~~~~~~~~~~~~~~~~~~~ In the `[jobsub]` section of the configuration file, you should specify: oozie_url:: The URL of the Oozie service. It is the same as the `OOZIE_URL` environment variable for Oozie. UserAdmin Configuration ~~~~~~~~~~~~~~~~~~~~~~~ In the `[useradmin]` section of the configuration file, you can _optionally_ specify the following: default_user_group:: The name of a default group that is suggested when creating a user manually. If the LdapBackend or PamBackend are configured for doing user authentication, new users will automatically be members of the default group. Starting Hue from the Tarball ----------------------------- After your cluster is running with the plugins enabled, you can start Hue on your Hue Server by running: # build/env/bin/supervisor Your Hue installation is now running. Administering Hue ----------------- Now that you've installed and started Hue, you can feel free to skip ahead to the <> section. Administrators may want to refer to this section for more details about managing and operating a Hue installation. Hue Processes ~~~~~~~~~~~~~ ==== Process Hierarchy A script called `supervisor` manages all Hue processes. The supervisor is a watchdog process -- its only purpose is to spawn and monitor other processes. A standard Hue installation starts and monitors the following processes: * `runcpserver` - a web server based on CherryPy that provides the core web functionality of Hue * `beeswax server` - a daemon that manages concurrent Hive queries If you have installed other applications into your Hue instance, you may see other daemons running under the supervisor as well. You can see the supervised processes running in the output of `ps -f -u hue`: ------------------ UID PID PPID C STIME TTY TIME CMD hue 8685 8679 0 Aug05 ? 00:01:39 /usr/share/hue/build/env/bin/python /usr/share/hue/build/env/bin/desktop runcpserver hue 8695 8679 0 Aug05 ? 00:00:06 /usr/java/jdk1.6.0_14/bin/java -Xmx1000m -Dhadoop.log.dir=/usr/lib/hadoop-0.20/logs -Dhadoop.log.file=hadoop.log ... ------------------ Note that the supervisor automatically restarts these processes if they fail for any reason. If the processes fail repeatedly within a short time, the supervisor itself shuts down. === Hue Logging The Hue logs are found in `/var/log/hue`. Inside the log directory you can find: * An `access.log` file, which contains a log for all requests against the Hue web server. * A `supervisor.log` file, which contains log information for the supervisor process. * A `supervisor.out` file, which contains the stdout and stderr for the supervisor process. * A `.log` file for each supervised process described above, which contains the logs for that process. * A `.out` file for each supervised process described above, which contains the stdout and stderr for that process. If users on your cluster have problems running Hue, you can often find error messages in these log files. If you are unable to start Hue from the init script, the `supervisor.log` log file can often contain clues. ==== Viewing Recent Log Messages through your Web Browser In addition to logging `INFO` level messages to the `logs` directory, the Hue web server keeps a small buffer of log messages at all levels in memory. You can view these logs by visiting `http://myserver:8888/logs`. The `DEBUG` level messages shown can sometimes be helpful in troubleshooting issues. === The Hue Database Hue requires a SQL database to store small amounts of data, including user account information as well as history of job submissions and Hive queries. By default, Hue is configured to use the embedded database SQLite for this purpose, and should require no configuration or management by the administrator. However, MySQL is the recommended database to use; this section contains instructions for configuring Hue to access MySQL and other databases. ==== Inspecting the Hue Database The default SQLite database used by Hue is located in: `/usr/share/hue/desktop/desktop.db`. You can inspect this database from the command line using the `sqlite3` program. For example: --------- # sqlite3 /usr/share/hue/desktop/desktop.db SQLite version 3.6.22 Enter ".help" for instructions Enter SQL statements terminated with a ";" sqlite> select username from auth_user; admin test sample sqlite> ---------- It is strongly recommended that you avoid making any modifications to the database directly using SQLite, though this trick can be useful for management or troubleshooting. ==== Backing up the Hue Database If you use the default SQLite database, then copy the `desktop.db` file to another node for backup. It is recommended that you back it up on a regular schedule, and also that you back it up before any upgrade to a new version of Hue. ==== Configuring Hue to Access Another Database Although SQLite is the default database type, some advanced users may prefer to have Hue access an alternate database type. Note that if you elect to configure Hue to use an external database, upgrades may require more manual steps in the future. The following instructions are for MySQL, though you can also configure Hue to work with other common databases such as PostgreSQL and Oracle. [NOTE] .Tested Database Backends ============================================================ Note that Hue has only been tested with SQLite and MySQL database backends. ============================================================ ===== Configuring Hue to Store Data in MySQL To configure Hue to store data in MySQL: 1. Create a new database in MySQL and grant privileges to a Hue user to manage this database. mysql> create database hue; Query OK, 1 row affected (0.01 sec) mysql> grant all on hue.* to 'hue'@'localhost' identified by 'secretpassword'; Query OK, 0 rows affected (0.00 sec) 2. Shut down Hue if it is running. 3. To migrate your existing data to MySQL, use the following command to dump the existing database data to a text file. Note that using the ".json" extension is required. $ /usr/share/hue/build/env/bin/hue dumpdata > .json 4. Open the `/etc/hue/hue.ini` file in a text editor. Directly below the `[[database]]` line, add the following options (and modify accordingly for your MySQL setup): host=localhost port=3306 engine=mysql user=hue password=secretpassword name=hue 5. As the Hue user, configure Hue to load the existing data and create the necessary database tables: $ /usr/share/hue/build/env/bin/hue syncdb --noinput $ mysql -uhue -psecretpassword -e "DELETE FROM hue.django_content_type;" $ /usr/share/hue/build/env/bin/hue loaddata .json Your system is now configured and you can start the Hue server as normal. [[usage]] == Using Hue After installation, you can use Hue by navigating to `http://myserver:8888/`. The following login screen appears: image:images/login.png[] Launch applications on the bottom-right. image:images/open-apps.png[,width=50%] The Help application guides users through the various installed applications. === Supported Browsers Hue is primarily tested on Firefox 3.5 and Firefox 3.6, on Windows, Mac, and Linux. Google Chrome and Safari work as well. === Feedback Your feedback is welcome. The best way to send feedback is to join the https://groups.google.com/a/cloudera.org/group/hue-user[mailing list], and send e-mail, to mailto:hue-user@cloudera.org[hue-user@cloudera.org]. === Reporting Bugs If you find that something doesn't work, it'll often be helpful to include logs from your server. These are available at the +/logs+ URL on Hue's web server (not part of the graphical Hue UI). Please download the logs as a zip (or cut and paste the ones that look relevant) and send those with your bug reports. image:images/logs.png[] ==== Starting Hue To start Hue, use `build/env/bin/supervisor`. This will start several subprocesses, corresponding to the different Hue components.