You can install Streams Studio on a local Linux or Windows system that acts as a client and install IBM®
Streams on a remote Linux system that acts as a server. To connect
to the remote Linux system where IBM
Streams is installed, you must configure a connection.
Before you begin
Install Streams Studio on your local Linux or Windows system.
Ensure that you have a user account on the remote Linux
system where IBM
Streams is installed.
Streams Studio uses Remote System Explorer (RSE) to provide connectivity to the remote Linux system for building streams processing applications. When you run
Streams Studio for the first time on the local Linux or Windows system, you must configure a connection to specify how you want to
start the RSE server on the remote system and establish a connection. You have two options for
configuring a connection:
- Create a connection configuration file by editing the default.studioconfig
sample file in your Streams Studio installation directory. When you start Streams Studio in the following procedure, the configuration values in this file are automatically entered for
you.
- Specify the configuration values when you complete the following procedure.
If Streams Studio has problems accessing the IBM
Streams domain, see the firewall configuration guidelines for clients that are external to the IBM
Streams cluster.
Procedure
- Start Streams Studio on your local system.
- On a Linux system, go to the directory where you
installed Streams Studio and enter ./streamsStudio.
- On a Windows system, go to the directory where you
installed Streams Studio and double-click streamsStudio.exe.
- In the Workspace Launcher window, specify the local workspace where you
want to store your SPL projects, and then click OK. This workspace is a directory on the local system where you installed Streams Studio.
- In the IBM Streams Install Location Details window:
- If you are using a connection configuration file, the connection to the remote Linux system is automatically created and displayed in the
Host connection field. This connection is based on the values in your
.studioconfig file.
- If you are not using a connection configuration file, click New to
create a connection to the remote Linux system.
- In the Select Remote System Type window, select the remote Linux system type, and then click Next.
- In the New Connection window, specify the following values, and then click
Next.
- Parent profile name. RSE creates a default profile that typically uses the name of the local
system. Accept the default profile or specify an RSE profile of your choice.
- Host name or IP address of the remote Linux system where
IBM
Streams is installed
- Connection name that is unique to your profile. This name will be displayed in the tree view in
the Remote Systems view.
- Optional description
- Optional proxy settings for the remote Linux system
- In the New remote connection configuration window, configure the connection
properties to define how you want to connect to the RSE server, and then click
Finish.
- Remote daemon option: RSE establishes the connection by using the remote
daemon. The remote daemon must be running on the remote Linux
system at a predefined port and must be started by a root user.
Enter the following commands to
start the remote
daemon:
su -l root
cd product-installation-root-directory/4.2.1.0/etc/rseserver
perl ./daemon.pl [daemonPort] [serverPortRange]
where
product-installation-root-directory is the root directory
for the IBM
Streams product
installation.Notes: - By default, the server daemon runs on port 8050. To use a different port, you can specify the
optional daemonPort argument. If your daemon runs behind a firewall, you might want
to use the optional serverPortRange argument to restrict selected server ports to
the range that you specify. For example, perl ./daemon.pl 4075 10000-10010
specifies a port range of 10000-10010.
- The server daemon uses the Perl script auth.pl to authenticate the user who
is making the connection. The auth.pl script uses basic password authentication
by using the password file. You must update the script to use the authentication mechanism that
matches your system.
- SSH option: RSE establishes the connection by using Secure Shell (SSH)
support. To establish a connection, the SSH command, ./server.sh, is issued to
start the RSE server. You must specify the path where the RSE server is installed on the remote Linux system.
Note: If your remote Linux system is behind a firewall, you might want to restrict the server to
use a specified port range by setting the port variable in the server.sh file.
To set the port variable, edit the server.sh file on your remote Linux system and set the port variable to the desired port range.
For example, port=10000-10010; specifies a port range of
10000-10010.
In the
Path to installed server on host field, specify
an absolute path or a path relative to your home directory, as shown in the following
examples:
/home/user-directory/product-installation-root-directory/4.2.1.0/etc/rseserver
product-installation-root-directory/4.2.1.0/etc/rseserver
- When prompted, enter your user ID and password.
- To specify the location where IBM
Streams is installed, click Browse and select the directory where IBM
Streams is installed on the remote Linux system.
Note: If this information is specified in the connection configuration file, the value is already
entered.
- To specify the remote workspace, click Browse, expand My
Home, and then select a folder on the remote Linux system. To create a folder on the remote Linux system from your
local system, select the parent folder, and click .
Results
The
Remote Systems view displays the short name of the connection with
the following nodes under the connection:
- Files
- Processes
- Contexts
- Shells
- Ssh Terminals
What to do next
After establishing the connection, you can access the file system and run remote commands
on the remote Linux system. You can import, create, build,
run, and monitor streams processing applications; create
IBM
Streams runtime instances; and view instance and application graphs.