Home
Installing
An HCL Launch installation consists of the HCL Launch server, a database, and at least one agent.
Starting HCL Launch
Both UNIX™ and Windows™ installations require the HCL Launch server and at least one agent. If you are using a database other than Derby, make sure that the appropriate driver is installed and configured.
Starting z/OS agents from the z/OS UNIX command line
The z/OS agent can be started from ISPF as a started task or from the z/OS UNIX command line. To start an agent from ISPF as a started task refer to install/topics/run_agent.html. To start the z/OS agent from the z/OS UNIX command line, follow this procedure.

Installing
An HCL Launch installation consists of the HCL Launch server, a database, and at least one agent.
- System requirements and performance considerations
  The HCL Launch server runs on Windows™ and UNIX™ systems. Following some guidelines can improve the performance of the server and agents.
- Firewall and communication configuration
  This topic describes the communication between the HCL Launch components, the firewall, and the ports involved in the communication.
- Installing the server database
  HCL Launch requires a database. You can use the following database types for an evaluation or production server.
- Installing the server
  The server provides services such as the user interface used to configure application deployments, the workflow engine, the security service, and the artifact repository, among others. Installing the server involves specifying configuration information for the server, such as the ports that the server uses.
- Installing with Kubernetes
  The HCL Launch server can be installed in a cloud-based environment, as a containerized server cluster, managed by Kubernetes. HCL Launch works with Kubernetes to simplify application deployment and manage versions in the containers.
- Installing agents
  You can install agents from the server or from the command line on the agent system. To configure a failover server for an agent, you must install the agent from the command line.
- Installing agent relays
  An agent relay is a communication proxy for agents that are located behind a firewall or in another network location. You can install agent relays in interactive mode or in silent mode.
- Installing plug-ins
  Plug-ins can be installed at any time. You do not need to restart the server after you install a plug-in.
- Starting HCL Launch
  Both UNIX™ and Windows™ installations require the HCL Launch server and at least one agent. If you are using a database other than Derby, make sure that the appropriate driver is installed and configured.
  - Starting and stopping the server
    To run the server, run the batch file or shell script for the server.
  - Starting agents
    To start an agent, run the shell script or batch file.
  - Starting z/OS agents from the z/OS UNIX command line
    The z/OS agent can be started from ISPF as a started task or from the z/OS UNIX command line. To start an agent from ISPF as a started task refer to install/topics/run_agent.html. To start the z/OS agent from the z/OS UNIX command line, follow this procedure.
  - Starting agent relays
    To start an agent relay, run its batch file or shell script.
  - Accessing HCL Launch
    You can access the HCL Launch server with a web browser.
- Installing and uninstalling test fixes (patches) from the server
  Test fixes are provided to correct specific issues. They are applied to the server as patches.

Starting z/OS agents from the z/OS UNIX command line

The z/OS agent can be started from ISPF as a started task or from the z/OS UNIX command line. To start an agent from ISPF as a started task refer to Starting agents. To start the z/OS agent from the z/OS UNIX command line, follow this procedure.

Before you begin

Make sure that the server is running. Also, if the agent runs through an agent relay, make sure that the relay is started.

Procedure

Login to the z/OS UNIX shell using the TSO OMVS command or by using a TELNET client.
Go to the agent_install_dir/bin directory.
Run the following script:
```
./agent start
```
To stop the agent, run the following script from the same directory:
```
./agent.stop
```

Results

When the agent is started from the command line, the current login user ID is used as the agent user ID.