Instructions for both Linux and Windows systems are included in this section.
Contents
In addition to meeting the recommended system requirements, the server on which you wish to install Metacat must have the following software installed and running correctly:
System requirements for running Metacat:
This section contains instructions for downloading and installing Metacat on Linux systems. As Mac OS X is based on BSD Unix, these Linux instructions can be adapted to also work on Mac OS X (although the exact commands for downloading and installing packages will differ due to the different package management approaches on the Mac).
For the impatient or those who have already installed Metacat and know what they are doing, here are the steps needed to install Metacat. Detailed instructions for each step are in the next section.
pb_hba.conf
for the user ‘metacat’sudo mkdir /var/metacat; sudo chown -R <tomcat_user> /var/metacat
sudo cp <metacat_package_dir>/metacat.war <tomcat_app_dir>
sudo /etc/init.d/tomcat7 restart
Before installing Metacat, please ensure that all required software is installed and running correctly. To obtain a Metacat WAR file, which is needed for installation, download one of the following:
Instructions for all three options are discussed below. Note that downloading the installer (described in the next section) is the simplest way to get started.
Downloading the Metacat Installer is the simplest way to get started with the application. To download the installer:
tar -xvzf metacat-bin-X.X.X.tar.gz
You should see a WAR file and several sample supporting files (Table 2.1). The
extraction location will be referred to as the <metacat_package_dir>
for the
remainder of this documentation.
File | Description |
---|---|
metacat.war | The Metacat Web archive file (WAR) |
metacat-site.conf | Sample Web definition file used by Apache on Ubuntu/Debian Linux systems. |
metacat-site-ssl.conf | Sample SSL definition file used by Apache on Ubuntu/Debian Linux systems. |
jk.conf | Sample JkMount configuration file used by Apache on Ubuntu/Debian Linux systems. |
workers.properties | Sample workers definition file used by Apache on Ubuntu/Debian Linux systems. |
metacat-index.war | The Metacat Index WAR for supporting SOLR query features Optional unless Metacat UI is being used. |
metacatui.war | The Metacat UI - can be deployed as a webapp or directly in webserverMetacat UI requires metacat-index be deployed and configured. |
authority.war | The optional LSID Server application WAR |
To get the Metacat source distribution:
tar -xvzf metacat-src-X.X.X.tar.gz
Note that you do not need to create the WAR file directly because the Ant build-file has an “install” target that will build and deploy the WAR for you.
If you wish to work with the most recent Metacat code, or you’d like to extend the Metacat code yourself, you may wish to check out the Metacat source code from SVN. You will need a Subversion (SVN) client installed and configured on your system (see the end of this section for information about obtaining an SVN client).
To check out the code from SVN, go to the directory where you would like the code to live and type:
svn co https://code.ecoinformatics.org/code/metacat/tags/METACAT_<rev> metacat
Where <rev>
is the version of the code you want to check out (like 2_0_0).
To check out the head, type:
svn co https://code.ecoinformatics.org/code/metacat/trunk metacat
You should see a list of files as they check out.
Note that you do not need to create the WAR file directly because the Ant build-file has an “install” target that will build and deploy the WAR for you.
Before you can install and run Metacat, you must ensure that a recent Java SDK, PostgreSQL, Ant (if installing from source), and Tomcat are installed and running correctly. We also highly recommend that you install Apache Web server, as it provides a more robust Web-serving environment and is required by some Metacat functionality.
To run Metacat, you should use Java 7. Make sure that the JAVA_HOME
environment variable is properly set and that both java
and javac
are on your PATH.
To install Java if you are running Ubuntu/Debian, you can install using apt-get:
sudo apt-get install openjdk-7-jdk
If you are not using Ubuntu/Debian, you can get Java from the Oracle website and install using the RPM or other installer (Windows).
We recommend that you install Tomcat 6 or 7 into the directory of your choice. Included with the Metacat download is a Tomcat-friendly start-up script that should be installed as well.
Note: we will refer to the Tomcat installation directory as <tomcat_home>
for
the remainder of the documentation.
If you are running Ubuntu/Debian, get Tomcat by typing:
sudo apt-get install tomcat7
Otherwise, get Tomcat from the Apache Tomcat page.
After installing Tomcat, you can switch back to the Sun JDK by typing:
sudo update-alternatives --config java
and selecting the correct Java installation.
If using Tomcat with Apache/mod_jk, enable the AJP connector on port 8009 by uncommenting that section in:
<tomcat_home>/conf/server.xml
For DataONE deployments edit:
/etc/tomcat7/catalina.properties
to include:
org.apache.tomcat.util.buf.UDecoder.ALLOW_ENCODED_SLASH=true
org.apache.catalina.connector.CoyoteAdapter.ALLOW_BACKSLASH=true
Although you have the option of running Metacat with only the Tomcat server, we highly recommend that you run it behind the Apache Web server for several reasons; running Tomcat with the Apache server provides a more robust Web serving environment. The Apache Web server is required if you wish to install and run the Metacat Registry or to use the Metacat Replication feature.
This section contains instructions for installing and configuring the Apache Web server for Metacat on an Ubuntu/Debian system. Instructions for configuring Apache running on other Linux systems are included in the sidebar.
sudo apt-get install apache2 libapache2-mod-jk
If you are installing the Apache server on an Ubuntu/Debian system, and you installed Apache using apt-get as described above, the Metacat code will have helper files that can be dropped into directories to configure Apache. Depending on whether you are installing from binary distribution or source, these helper files will be in one of two locations:
<metacat_code_dir>/src/scripts
(for both the source distribution and source code checked out from SVN). We will refer to the directory with the helper scripts as <metacat_helper_dir>
and the directory where Apache is installed (e.g., /etc/apache2/
) as <apache_install_dir>
.sudo cp <metacat_helper_dir>/debian/jk.conf <apache_install_dir>/mods-available
sudo cp <metacat_helper_dir>/debian/workers.properties <apache_install_dir>
sudo a2dismod jk
sudo a2enmod jk
sudo cp <metacat_helper_dir>/metacat-site.conf <apache_install_dir>/sites-available
sudo a2ensite metacat-site.conf
sudo a2dissite 000-default
sudo /etc/init.d/apache2 restart
Currently Metacat only supports PostgreSQL. To install and configure PostgreSQL:
sudo apt-get install postgresqlOn other systems, install the rpms for postgres.
sudo /etc/init.d/postgresql-8.4 start
sudo su - postgres
gedit /etc/postgresql/8.4/main/pg_hba.confAdd the following line to the configuration file:
host metacat metacat 127.0.0.1 255.255.255.255 passwordSave the file and then create the Metacat instance:
createdb metacat
psql metacat
CREATE USER metacat WITH UNENCRYPTED PASSWORD 'your_password';where ‘your_password’ is whatever password you would like for the Metacat user.
\q
/etc/init.d/postgresql-8.4 restart
logout
psql -U metacat -W -h localhost metacat
\q
The Metacat servlet automatically creates the required database schema. For more information about configuring the database, please see Database Configuration.
If you are building Metacat from a source distribution or from source code checked out from SVN, Ant is required. (Users installing Metacat from the binary distribution do not require it.) Ant is a Java-based build application similar to Make on UNIX systems. It takes build instructions from a file named “build.xml”, which is found in the root installation directory. Metacat source code comes with a default “build.xml” file that may require some modification upon installation.
If you are running Ubuntu/Debian, get Ant by typing:
sudo apt-get install ant
Otherwise, get Ant from the Apache Ant homepage.
Ant should be installed on your system and the “ant” executable shell script should be available in the user’s path. The latest Metacat release was tested with Ant 1.8.2.
Instructions for a new install, an upgrade, and a source install are included below.
Before installing Metacat, please ensure that all required applications are installed, configured to run with Metacat, and running correctly. If you are upgrading an existing Metacat servlet, please skip to Upgrade. For information about installing from source, skip to Source Install and Upgrade.
To install a new Metacat servlet:
sudo mkdir /var/metacat
sudo chown -R <tomcat_user> /var/metacat
sudo cp <metacat_package_dir>/metacat.war <tomcat_app_dir>
sudo /etc/init.d/tomcat7 restart
Congratulations! You have now installed Metacat. If everything is installed correctly, you should see the Authentication Configuration screen (Figure 2.1) when you type http://yourserver.com/yourcontext/ (e.g., http://knb.ecoinformatics.org/knb) into a browser. For more information about configuring Metacat, please see the Configuration Section.
To upgrade an existing binary Metacat installation follow the steps in this section. The steps for upgrading Metacat from source are the same as the instructions for installing from source:
/etc/init.d/tomcat7 stop
cp <web_app_dir>/metacat <backup_dir>/metacat.<yyyymmdd> cp <web_app_dir>/metacat.war <backup_dir>/metacat.war.<yyyymmdd>Warning: Do not backup the files to the
<web_app_dir>
directory. Tomcat will try to run the backup copy as a service.
sudo cp <metacat_package_dir>/metacat.war <tomcat_app_dir>Note: Typically, Tomcat will look for its application files (WAR files) in the
<tomcat_home>/webapps
directory. Your instance of Tomcat may be configured to look in a different directory.
sudo cp <metacat_package_dir>/authority.war <tomcat_app_dir>
/etc/init.d/tomcat7 restart
7. Run your new Metacat servlet. Go to a Web browser and visit your installed Metacat application, using a URL of the form:
http://yourserver.yourdomain.com/yourcontext/
You should substitute your context name for “yourcontext” in the URL above (your context will be “metacat” unless you change the name of the metacat.war file to something else). If everything is working correctly, you should be presented with Metacat’s Authorization Configuration screen. Note that if you do not have Tomcat integrated with Apache you will probably have to type http://yourserver.yourdomain.com:8080/yourcontext/
Whether you are building Metacat from the source distribution or source code checked out from SVN, you will need Apache Ant to do the build (see Installing and Configuring Required Software for more information about Ant).
To install Metacat from source:
<metacat_src_dir>
.
- Set the build.tomcat.dir property to your Tomcat installation directory. Metacat will use some of the native Tomcat libraries during the build. For instance: build.tomcat.dir=/usr/local/tomcat
- Set the app.deploy.dir property to your application deployment directory. For instance: app.deploy.dir=/usr/local/tomcat/webapps
<metacat_src_dir>
, run:sudo ant clean installYou will see the individual modules get built. You should see a “BUILD SUCCESSFUL” message at the end.
You should see a new file named metacat.war in your application deployment directory.
To run your new Metacat servlet, open a Web browser and type:
http://yourserver.yourdomain.com/yourcontext/
(e.g. http://knb.ecoinformatics.org/metacat/)
Your context will be “metacat” unless you changed the name of the metacat.war file to something else. The servlet may require a few seconds to start up, but once it is running, you will be presented with the Authorization Configuration screen.
Note
The support for LSID identifiers is deprecated, and is being replaced with support for DOI identifiers in a future release. We are maintaining support for LSIDs on one particular site, but this support will be removed in a future version of Metacat.
Metacat’s optional LSID server allows Metacat to use a standardized syntax for identifying data sets, in addition to Metacat’s internal, custom scheme for identifiers. LSID’s were designed to identify complex biological entities with short identifiers (much like DOIs in publishing) that are both computer and human readable. LSID identifiers are URIs and are therefore usable in many Internet applications, but they also cleanly separate the identity of a data set (i.e., its permenant identifier) from its current location (e.g., the list of URLs from which it might be retrieved). LSIDs accomplish this by using a level of indirection; the identifier represents simply a name without location, but an associated resolver service can be used to locate the current location of the data and medata for the data set. This is accomplished by establishing a well-known location for the resolution service for each authority using an infrequently used feature of the domain name system called SRV records. At its most basic, resolution of an identifier is performed when a client looks up the SRV record for an LSID by querying DNS, which returns the current host and port of the authority web service, which is in turn used to locate the data and metadata.
Using LSIDs to identify data records is being debated among members of the Taxonomic Databases Working Group (TDWG). There are several alternate technologies that are under consideration (e.g., DOI, plain http URIs), and so at this time the support for LSIDs in Metacat has been created on an experimental basis only. If the LSID approach is ratified by the broader community, we will expand support for LSIDs in Metacat, but until then it is an optional and experimental feature.
The format of an LSID is:
urn:lsid:<Authority>:<Namespace>:<ObjectID>[:<Version>]
e.g., urn:lsid:ecoinformatics.org:tao:12039:1
When you enable the Metacat LSID support, you can use LSID clients (such as LSID Launchpad) and LSID notation to query Metacat for data and metadata. LSID notation can be used directly in Metacat HTTP queries as well. For example, a data package with an ID tao.12039.1 that is stored in a Metacat available at: http://example.com:9999 can be accessed by the following HTTP Metacat queries:
http://example.com:9999/authority/data?lsid=urn:lsid:ecoinformatics.org:tao:12039:1
(To return the data)
http://example.com:9999/authority/metadata?lsid=urn:lsid:ecoinformatics.org:tao:12039:1
(To return the metadata)
Notice that in the HTTP query strings, the periods in the data package ID have been replaced with colons. The authority (ecoinformatics.org) must be properly configured by the Metacat administrator. Note: In order to configure the authority, you must have access to the DNS server for the Metacat domain. Further instructions are provided below.
To install the LSID server using the binary installation:
sudo cp <metacat_package_directory>/authority.war /usr/share/tomcat7/webapps
Set up the LSID server by dropping the authority file into Apache’s sites-available directory and running a2ensite to enable the site:
sudo cp <metacat_helper_dir>/authority /etc/apache2/sites-available
sudo a2ensite authority
Restart Tomcat. Log in as the user that runs your Tomcat server (often “tomcat”) and type:
/etc/init.d/tomcat5.5 restart
Restart Apache to bring in changes by typing:
sudo /etc/init.d/apache2 restart
authority.context=authority config.lsidauthority=ecoinformatics.org
sudo ant war-lsid
sudo cp <metacat_package_dir>/dist/authority.war <tomcat_app_dir>
/etc/init.d/tomcat7 restart
sudo cp <metacat_helper_dir>/authority <apache_install_dir>/sites-available sudo a2ensite authorityWhere <metacat_helper_dir> can be found in <metacat_code_dir>/src/scripts
sudo /etc/init.d/apache2 restartOnce the authority.war is installed, you must also modify the SRV record(s) on the DNS server for the domain hosting the Metacat. The record should be added to the master zone file for the metacat’s DNS server:
_lsid._tcp IN SRV 1 0 8080 <metacat.edu>.Where <metacat.edu> is the name of the machine that will serve as the physical location of the AuthorityService.
For example, the value of <metacat.edu> for the below example URL would be example.com:
http://example.com:9999/authority/data?lsid=urn:lsid:ecoinformatics.org:tao:12039:1For more information, please see http://www.ibm.com/developerworks/opensource/library/os-lsid/
We keep and update a list of common problems and their solutions on the KNB website. See http://knb.ecoinformatics.org/software/metacat/troubleshooting.html for more information.
Metacat can be installed on Windows. Please follow the instructions in this section for downloading Metacat, installing the required software, and installing Metacat. Note that Registry and Data Upload functionality has not been tested on Windows.
To obtain a Metacat WAR file, which is used when installing the Metacat servlet:
Note: The location where these files were extracted will be referred to as the
<metacat_package_dir>
for the remainder of this documentation.
Note: Before installing Metacat, please ensure that all required software is installed and running correctly.
Before you can install and run Metacat, you must ensure that a recent Java SDK, PostgreSQL and Tomcat are installed, configured, and running correctly.
To run Metacat, you must have Java 7. Make sure that the JAVA_HOME environment variable is properly set and that both java and javac are on your PATH.
To download and install Java:
System Variable: JAVA_HOME C:\Program Files\Java\jdk1.7.0_79 (or whichever version you downloaded)
We recommend that you install Tomcat version 7. To download and install Tomcat:
Metacat only can be run with PostgreSQL. Instructions for installing and configuring PostgreSQL for use with Metacat are included in this section.
To download and install PostgreSQL:
host metacat metacat 127.0.0.1 255.255.255.255 password
C:\Program Files\PostgreSQL\8.3\bin createdb -U postgres metacat (enter super user password)
psql -U postgres metacat (enter super user password)
CREATE USER metacat WITH UNENCRYPTED PASSWORD 'your_password'
\q
run start/All Programs/PostgreSQL 8.3/Stop Server run start/All Programs/PostgreSQL 8.3/Start Server
psql -U metacat -W -h localhost metacat
\q
The Metacat servlet automatically creates the required database schema. For more information, please see Database Configuration.
Instructions for a new install and for an upgrade are included below.
Before installing Metacat, please ensure that all required applications are installed, configured to run with Metacat, and running correctly. If you are upgrading an existing Metacat servlet, please skip to Upgrade.
To install a new Metacat servlet:
C:/Program Files/metacat
copy <metacat_package_dir>\metacat.war C:\Program Files\tomcat\webapps
C:\Program Files\tomcat\shutdown.bat C:\Program Files\tomcat\startup.bat
Congratulations! You are now ready to configure Metacat. Please see the Configuration Section for more information.
To upgrade an existing Metacat installation:
<web_app_dir>/metacat to <backup_dir>/metacat.<yyyymmdd> <web_app_dir>/metacat.war to <backup_dir>/metacat.war.<yyyymmdd>Warning: Do not backup the metacat directory in the <web_app_dir> directory. Tomcat will try to run the backup copy as a service.
copy metacat.war C:\Program Files\tomcat\webapps
C:\Program Files\tomcat\shutdown.bat C:\Program Files\tomcat\startup.bat
Congratulations! You are now ready to configure Metacat. Please see Configuring Metacat for more information.