Page History

{scrollbar}
{excerpt}The *deploy* script is used for installing, uninstalling, reinstalling, starting and stopping applications and modules{excerpt} and for installing and uninstalling configurations (for example some configuration specific deployment plans, security realms, database connection pools etc.)

h1. Usage

This command has the following syntax:

{noformat}
deploy <general_options> <command> <command_options>
{noformat}
where _<general_options>_ specify common options that apply to all commands and control how the application behaves, _<command>_ is a command name that specifies the action to be performed, and _<command_options>_ are options unique to the command specified.

The *deploy* command can also be started by using the java \-jar command:
{noformat}
java -Djava.endorsed.dirs=lib/endorsed -jar bin/deployer.jar <general_options> <command> <command_options>
{noformat}

h2. General options

This section lists all the available general options for the Geronimo deployer tool.
* *\--uri* <identifier>
Where _<identifier>_ is a Universal Resource Identifier (URI) that specifies how the deployer is to contact the server. If this flag is not specified, the deployer will attempt to contact the server using the standard port on localhost. The identifier must have the following form:
{noformat}
deployer:geronimo:jmx:rmi:///jndi/rmi://host:port/JMXConnector
{noformat}
where _<host>_ is replaced with the host name or TCP/IP address of the system where the server is running and <port> is replaced with the port number where the server is listening. If unspecified, localhost and the default port will be used.

* *\--host* <host>
Where _<host>_ is the host name of the server you are trying to deploy that application or resource. This option allows you to deploy resources and applications to a remote server. This parameter is optional and defaults to localhost defined as RemoteDeployHostname=localhost in <Geronimo_home>/var/config/config-substitutions.properties. Change localhost to the server's ip address if you want remote-deployment enabled on it.

* *\--port* <port>
Where _<port>_ is the port of the remote server you are trying to deploy that application or resource. This parameter is optional and defaults to port 1099.

* *\--driver* <driver_path>
Where _<driver_path>_ is the path to the driver JAR if you want to use this tool with a server other than Geronimo. Currently, manifest Class-Path entries in that JAR are ignored.

* *\--user* <username>
Where _<username>_ is a user name authorized to be an administrator on the server. If the command requires authorization, you must use this option.

* *\--password* <password>
Where _<password>_ is a the password required to authenticate the user name. If this flag is not specified, the deployer will attempt to perform the command with no password, but if that fails, it will prompt you to enter a password.

* *\--secure*
Use secure channel to communicate with a JMX server, see [#Security] for details. 

* *\--syserr* <select>
Where _<select>_ can be either true or false. If this flag is unspecified. false is assumed. Specify true when you want errors to be logged to the syserr device.

* *\--verbose* <select>
Where _<select>_ can be either true or false. If this flag is unspecified. false is assumed. Specify true when you need more messages to determine the cause of an error.

[Back to top|#top]

h2. Security

Starting with Geronimo 2.1.2, the deployer tool can use a secure channel (SSL/TLS) to communicate with the JMX server to perform the given actions. To enable secure communication just add the *\--secure* option. Depending on your configuration you might also need to specify some Java security properties to configure the JVM to use the right keystores and passwords. For example, on a default Geronimo installation you might need to set the following (all in one line):

{noformat}
export JAVA_OPTS="-Djavax.net.ssl.trustStore=<geronimo_home>/var/security/keystores/geronimo-default -Djavax.net.ssl.trustStorePassword=secret"
{noformat}

Once that property is set, you can execute the following command (just as an example):

{noformat}
<GERONIMO_HOME>/bin/deploy -u system -p manager --secure list-modules
{noformat}

{note}The secure JMX server is not running by default. Please see [Securing RMI port of Geronimo] for more information.{note}
[Back to top|#top]

h2. Sub-commands

The available sub-commands for the Geronimo deployer tool are listed below:
* Common commands
** [#Deploy]
** [#Login]
** [#Redeploy]
** [#Start]
** [#Stop]
** [#Undeploy]
* Other commands
** [#Distribute]
** [#Encrypt]
** [Install-library|#Install-library]
** [List-modules|#List-modules]
** [List-targets|#List-targets]
** [#UnlockKeystore]
* Geronimo Plugins
** [Install-plugin|#Install-plugin]
** [Search-plugins|#Search-plugins]
* Bundle commands
** [#Install-bundle]
** [#Uninstall-bundle]

Additionally, you can type *help* for further details on a given command, the syntax is as follows:

{noformat}
deploy help <command>
{noformat}

[Back to top|#top]

h3. Deploy

Use the *deploy* command to add and start a new module. The deploy command has the following syntax:

{noformat}
deploy <general_options> deploy <module> <deployment_plan>
{noformat}

The <module> specifies the application file name and location. The <deployment_plan> specifies the file name and location of the XML with the deployment plan. Sometimes the application module already has included in the package a deployment plan or the application is so simple that does not require any deployment plan, in these cases this parameter can be omited.

A module file can be one of the following:
* J2EE Enterprise Application Archive (EAR) file
* J2EE Web Application Archive (WAR) file
* J2EE Enterprise JavaBean Archive (JAR) file
* J2EE Java Resource Archive (RAR) file

If the server is not currently running at the time of deploying the application, the module will be marked to start next time the server is started.

The most common <general_options> would be \--user and \--password. The \--inPlace option allows you point to and deploy an application directly from a directory external to Geornimo without the need for even packaging the application. In other words, you can have an application *running* in Geronimo but that application may be anywhere else on the file system.

{note}
Please note that the \--inPlace option cannot be used when deploying an application to a remote server.
{note}

To use this option you should type:

{noformat}
deploy <general_options> deploy --inPlace <APP_HOME>
{noformat}

Where <APP_HOME> indicates the home directory where you have your application (exploded).

You can also deploy applications if Geronimo is not running by using the \--offline option, the syntax for this command would be:

{noformat}
deploy <general_options> --offline deploy <module>
{noformat}

Off course, you can also combine \--offline and \--inPlace
{noformat}
deploy <general_options> --offline deploy --inPlace <APP_HOME>
{noformat}

[Back to top|#top]

h3. Login

Use the *login* command to save the username and password for the current connection to the file {{*.geronimo-deployer{*}}} in the current user's home directory. Future connections to the same server will try to use this saved authentication information instead of prompting where possible.

This information will be saved separately per connection URL, so you can specify \--url or \--host and/or \--port on the command line to save a login to a different server.

The *login* command has the following syntax:
{noformat}
deploy --user <user_name> --password <password> login
{noformat}

So, next time you run a different command that originally required user name and password, you can run the command directly, for example:

{noformat}
deploy list-modules
{noformat}

Similarly, you don't have to specify or re-enter the user name and password when you shut down the server using the [shutdown] command.

{warning}Even when the login information is not saved in clear text, it is not secure either. If you want to save the authentication securely, you should change the .geronimo-deployer file in your home directory so that nobody else can read or write it.
{warning}
[Back to top|#top]

h3. Redeploy

Use the *redeploy* command to stop, replace and restart a module that has been deployed before. The redeploy command has the following syntax:

{noformat}
deploy <general_options> redeploy -targets target<module> <deployment_plan>
{noformat}

Just like the [deploy command|#Deploy], the redeploy command accepts the following modules file types:
* J2EE Enterprise Application Archive (EAR) file
* J2EE Web Application Archive (WAR) file
* J2EE Enterprise JavaBean Archive (JAR) file
* J2EE Java Resource Archive (RAR) file

Typically, both a module and a plan are specified. If the module contains a plan or if a default plan can be used, the plan can be omitted. However, if a plan is specified in this case, it overrides the other plans. If the plan references a server component already deployed in the server's environment, the module is omitted.

Use *--targets* option only for clustering redeployment. For clustering redeployment you can find the  target with deploy [list-targets|#deploy-Listtargetscommandandoptions|List-targets command and options] command. Copy the one with the name as {{MasterConfigurationStore}} and use it as a target variable.
[Back to top|#top]

h3. Start

Use the *start* command to start a previously deployed module. The start command has the following syntax:

{noformat}
deploy <general_options> start <moduleIDs>
{noformat}

Where <moduleIDs> is a list of one or more modules (configID) separated by blank space. The module identification (or ConfigID) is defined at deployment time in the respective deployment plan for each module previously deployed.

[Back to top|#top]

h3. Stop

Use the *stop* command to stop a running module. The stop command has the following syntax:

{noformat}
deploy <general_options> stop <moduleIDs>
{noformat}

Where <moduleIDs> is a list of one or more modules (configID) separated by blank space. The module identification (or ConfigID) is defined at deployment time in the respective deployment plan for each module previously deployed.

[Back to top|#top]

h3. Undeploy

Use the *undeploy* command to stop and remove a module (running or not) and its deployment information from the server. The undeploy command has the following syntax:

{noformat}
deploy <general_options> undeploy <moduleIDs>
{noformat}

Where <moduleIDs> is a list of one or more modules (configID) separated by blank space. The module identification (or ConfigID) is defined at deployment time in the respective deployment plan for each module previously deployed.

This command has the same ability as with *deploy* to uninstall applications when the server is not running, this command has the following syntax:

{noformat}
deploy <general_options> --offline undeploy <moduleID>
{noformat}

[Back to top|#top]

h3. Distribute

Use the *distribute* command to add a new module to the server. This command does not start the module nor mark it to be started in the future. The distribute command has the following syntax:

{noformat}
deploy <general_options> distribute <module> <deployment_plan>
{noformat}

Just like with the deploy command, <module> specifies the application file name and location. The <deployment_plan> specifies the file name and location of the XML with the deployment plan. Sometimes the application module already has included in the package a deployment plan or the application is so simple that does not require any deployment plan, in these cases this parameter can be omitted.

A module file can be one of the following:
* J2EE Enterprise Application Archive (EAR) file
* J2EE Web Application Archive (WAR) file
* J2EE Enterprise JavaBean Archive (JAR) file
* J2EE Java Resource Archive (RAR) file

[Back to top|#top]

h3. Encrypt

Use the *encrypt* command to generate an encrypted string. This command takes use of&nbsp;org.apache.geronimo.util.EncryptionManager and has the following syntax:

{{*<geronimo_home>/bin/deploy <general_options> encrypt <string>*}}

Where <general_options> are common options that apply to all commands, <string> specifies a string to get encrypted.

Currently password strings are plain text in deployment plans, such as datasource or JMS deployment plans within an EAR. It might pose a security problem to store password strings as plain text even though the deployment plans are only used during the deployment process, and not at runtime. Starting from Geronimo 2.1.5, users can encrypt passwords using the encrypt command and paste the encrypted strings into deployment plans as password.

Examples:

Use this syntax to encrypt string {{passw0rd}} on an active server so that the encryption settings of that server will be used
{code}
deploy --user myadmin --password mypassword encrypt passw0rd
{code}

Online encryption result:
{code}
......
String to encrypt: passw0rd
Online encryption result:
{Simple}rO0ABXNyABlqYXZheC5jcnlwdG8uU2VhbGVkT2JqZWN0PjY9psO3VHACAARbAA1lbmNv
ZGVkUGFyYW1zdAACW0JbABBlbmNyeXB0ZWRDb250ZW50cQB+AAFMAAlwYXJhbXNBbGd0ABJMamF2YS9s
YW5nL1N0cmluZztMAAdzZWFsQWxncQB+AAJ4cHB1cgACW0Ks8xf4BghU4AIAAHhwAAAAEG2NoqXONCcU
GqfK0reVCpVwdAADQUVT
{code}

Use this syntax to encrypt string passw0rd offline
{code}
deploy --offline encrypt passw0rd
{code}

Offline encryption result:
{code}
......
String to encrypt: passw0rd
Offline encryption result:
{Simple}rO0ABXNyABlqYXZheC5jcnlwdG8uU2VhbGVkT2JqZWN0PjY9psO3VHACAARbAA1lbmNv
ZGVkUGFyYW1zdAACW0JbABBlbmNyeXB0ZWRDb250ZW50cQB+AAFMAAlwYXJhbXNBbGd0ABJMamF2YS9s
YW5nL1N0cmluZztMAAdzZWFsQWxncQB+AAJ4cHB1cgACW0Ks8xf4BghU4AIAAHhwAAAAEG2NoqXONCcU
GqfK0reVCpVwdAADQUVT
{code}

Note: Online encryption needs a running server to connect to and will use the encryption settings of that server, such as an encryption key, to do the encryption. As a result, the encrypted password usually can only be used for that particular server. Offline encryption uses the default encryption settings, and the encrypted password can be used by all servers. Offline encryption is thus less secure than online encryption.

[Back to top|#top]

h3. Install-library

Use the *install-library* command to install a library into server's repository. The install-library command has the following syntax:

{noformat}
deploy <general_options> install-library --groupId groupName <libraryFile>
{noformat}

Use the *\--groupId* option to specify a non-default group id for the library. Otherwise, the library file will be installed with the group id named *default*.

Examples:

{noformat}
deploy -u system -p manager install-library mylib-1.0.jar
{noformat}

That command will install the mylib-1.0.jar at <geronimo_home>/repository/default/mylib/1.0/mylib-1.0.jar

{noformat}
deploy -u system -p manager install-library --groupId mygroup mylib-1.0.jar
{noformat}

That command will install the mylib-1.0.jar at <geronimo_home>/repository/mygroup/mylib/1.0/mylib-1.0.jar

[Back to top|#top]

h3. List-modules

Use the *list-modules* command to list all available modules on the server, note that for running this command the server must be running. The list-modules command has the following syntax:

{noformat}
deploy <general_options> list-modules [--all|--started|--stopped]
{noformat}

* \--all : is used by default when no other option is specified. It will list all the available modules.
* \--started : this option will list only the modules that are running.
* \--stopped : this option will list only the modules that are not running.

[Back to top|#top]

h3. List-targets

Use the *list-targets* command to lists the targets known to the server you have connected to. The list-targets command has the following syntax:
{noformat}
deploy <general_options> list-targets
{noformat}

In the case of Geronimo, each configuration store is a separate target. Geronimo does not yet support clusters as targets.

[Back to top|#top]

h3. UnlockKeystore

Use the *unlockKeystore* command to unlock a keystore and private keys. The unlockKeystore command has the following syntax:

{{*<geronimo_home>/bin/deploy <general_options> unlockKeystore <keyStoreName> <keyAlias1> <keyAlias2>*}}

Where <keyStoreName> specifies a locked keystore to get unlocked, <keyAlias1> and <keyAlias2> are optionally used to specify one or more locked private keys in the keystore to get unlocked. 

Note that before you can use the unlockKeystore command, you need to ensure that the following lines are added to <geronimo_home>/var/config/config-substitutions.properties:
{code}
<keyStoreName>=<keyStoreEncryptedPassword>
<keyAlias1>=<keyAlias1EncryptedPassword>
<keyAlias2>=<keyAlias1EncryptedPassword>
...
{code}

Where 
* <keyStoreName> is the name of the keystore.
* <keyStoreEncryptedPassword> is the encrypted password for the keystore, which can be generated by using the *encrypt* command. When you copy and paste the generated encrypted password to <geronimo_home>/var/config/config-substitutions.properties, there should be no space in the encrypted password string.
* <keyAlias1>, <keyAlias2> are the names of the private keys in the keystore.
* <keyAlias1EncryptedPassword>, <keyAlias2EncryptedPassword> are the encrypted passwords for the private keys, which can also be generated by using the encrypt command.

Alternatively, you can create a key file to contain the passwords of the keystore and its private keys. Use {{org.apache.geronimo.keyStoreTrustStorePasswordFile}} property to specify the key file. See [Creating your keystorefile for SSL authentication|Creating your keystorefile for SSL authentication] for more detailed instructions. 

Examples:

Use this syntax to unlock the keystore whose name is mykeystore
{code}
deploy --user myadmin --password mypassword unlockKeystore mykeystore
{code}

Use this syntax to unlock the keystore whose name is mykeystore and the private key whose alias is key1
{code}
deploy --user myadmin --password mypassword unlockKeystore mykeystore key1
{code}

[Back to top|#top]

h3. Install-plugin

Use the *install-plugin* command to install a Geronimo plugin previously exported from a Geronimo server or downloaded from a repository. A Geronimo plugin can be an application, a configuration such data sources and drivers or a combination. The install-plugin command has the following syntax:
{noformat}
deploy install-plugin <plugin_file>
{noformat}

[Back to top|#top]

h3. Search-plugins

Use the *search-plugins* command to list all the Geronimo plugins available in a Maven repository. The search-plugins command has the following syntax:

{noformat}
deploy search-plugins <maven_repository_URL>
{noformat}

[Back to top|#top]

h3.install-bundle command and options

The {{install-bundle}} command installs and records an OSGi bundle file in the server, so that the installed bundle can be automatically started even after you cleaned the cache directory of the OSGi framework. Use below syntax:

{panel:borderStyle=solid}
h6. {color:black}deploy _generalOptions_ install-bundle --groupId _groupId_ --startLevel _number_ {norender}[--start]{norender}{color}

where 

{indent:0}*_generalOptions_*{indent}
{indent:1}are [common options|#deploy] that apply to all commands. Note that you can only run the {{install-bundle}} command on the local server.{indent}
{indent:0}*_groupId_*{indent}
{indent:1}is optionally used to specifie the groupId for installed bundle, the default value is *default*.{indent}
{indent:0}*_number_*{indent}
{indent:1}is optionally used to specifies the start level of installed bundle. The default value is 60.{indent}
{indent:0}--start{indent}
{indent:1}specifies whether to start the installed bundle automatically. The bundle will be in *Resolved* state if you did not specify this argument.{indent}
{panel}

In the server repository, all modules are stored in a maven-style structure directory. After an OSGi bundle is installed, its maven artifact name is _groupId/Bundle-SymbolicName/Bundle-Version/jar_.

[Back to top|#top]

h3. Uninstall-bundle command and options {anchor:Uninstallbundlecommandandoptions}

The {{uninstall-bundle}} command stops and removes an OSGi bundle and its deployment information from the server, and uses this syntax:

{panel:borderStyle=solid}
h6. {color:black}deploy _generalOptions_ uninstall-bundle _bundleId_{color}

where 

{indent:0}*_generalOptions_*{indent}
{indent:1}are [common options|#deploy] that apply to all commands. Note that you can only run the {{uninstall-bundle}} command on the local server.{indent}{indent}
{indent:0}*_bundleId_*{indent}
{indent:1}is the bundle Id of OSGi bundle to be removed completely.{indent}
{panel}

*Note:* The {{uninstall-bundle}} command does not support {norender}--offline{norender} option.
[Back to top|#top]