1
0
Fork 0
mirror of https://github.com/eclipse-cdt/cdt synced 2025-07-04 15:45:25 +02:00
cdt/rse/doc/org.eclipse.rse.doc.user/tasks/tbeginlinux.html

186 lines
9.1 KiB
HTML
Executable file

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=UTF-8"/>
<meta name="copyright" content="Copyright (c) IBM Corporation and others 2004, 2007. This page is made available under license. For full details see the LEGAL in the documentation book that contains this page."/>
<link rel="stylesheet" type="text/css" href="../book.css"/>
<title>Connecting to a remote Linux or UNIX server</title>
</head>
<body id="tbeginlinux">
<a name="tbeginlinux"><!-- --></a>
<h1 class="topictitle1">Connecting to a remote Linux or UNIX server</h1>
<p>The following documentation explains how to install the Linux or UNIX server
code, start the server daemon, and make a connection to a remote Linux
or UNIX server. Look <a href="tbeginwindows.html">here</a> for setting up
a server on <a href="tbeginwindows.html">Windows</a>.</p>
<div>
<div class="p">
<p><b>Prerequisites</b></p>
<p>To use the Remote System Explorer communications server daemon you need
to install Perl. Using the daemon helps eliminate some of the manual steps
when you connect to the server.</p>
<p><b>Installing the server code</b></p>
<div class="p">
<ol>
<li>
Find the package that contains the server.
The server code is usually packaged with the containing product and you should refer to that
product's documentation for finding and installing the server package.
The server is also available, however, on the Eclipse DSDP
<a href="http://download.eclipse.org/dsdp/tm/downloads/">Target Management download site</a>
as the package <b>rseserver-&lt;version&gt;-&lt;os&gt;.tar</b>. For example, rseserver-2.0-linux.tar contains the release 2.0 server
for Linux. There are servers for Linux, AIX, a generic Unix version that can be tailored to your particular flavor
of Unix, and an experimental Mac OS X version.</li>
<li>Ensure that Perl is installed.</li>
<li>Ensure that a Java Runtime Environment (JRE) version 1.4 or higher is installed.
An IBM, Sun or equivalent JRE is required; The gcj-based jvm shipped with most Linux
distributions does not work. If in doubt, run the command <br/><tt>java -version</tt> (see
below) and check if there is a reference to gcj. You can download a Sun JRE from
<a href="http://java.sun.com">http://java.sun.com</a>.</li>
<li>Create a directory where you want to install the server code. The remainder
of these instructions will assume the directory /opt/rseserver (suitable for
team sharing), but you are free to use any directory.</li>
<li>Upload the server package to this directory. You can use FTP.</li>
<li>Switch to the /opt/rseserver directory by typing:
<pre>cd /opt/rseserver</pre>
</li>
<li>Run the following command in the /opt/rseserver directory to extract the
server code from the package appropriate to your operating system. For linux this command is:
<pre>tar -xf rseserver-2.0-linux.tar</pre>
</li>
</ol>
</div>
<p><b>Starting the server</b></p>
<div class="p">You can start the RSE communications
server with the server daemon, or manually. Before starting the server, make
sure the Java command is in your path, you can do this by running the following
command:<pre>java -version</pre>
</div>
<div class="p">You should see something
similar to the following:<pre>java version "1.4.1"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.4.1)
Classic VM (build 1.4.1, J2RE 1.4.1 IBM build cxppc321411-20040301 (JIT enabled: jitc))</pre>
</div>
<div class="p">If
you receive a "command not found" error, then try creating a symbolic link
to the java command in /usr/bin by running the following command:<pre>ln -s /opt/IBMJava2-141/jre/bin/java /usr/bin/java</pre>
</div>
<div class="p"><b>To start the server with the server daemon</b>
<ol>
<li>Ensure that you are running using the root user ID. (If the daemon is
not run under root, it will be unable to authenticate connecting users.) Run
the following commands:<pre>su -l root
cd /opt/rseserver
perl ./daemon.pl [daemonPort] [serverPortRange]</pre>
</li>
</ol>
Note that the server daemon runs on port 4075 by default. You can pass the optional
daemonPort argument to force a different port if you want.<br/>
If your daemon runs behind a firewall, you may want to specify the optional
<i>serverPortRange</i> argument to restrict selected server ports to the
range given:<pre>
perl ./daemon.pl 4075 10000-10010</pre>
<p><b>To start the server manually</b></p>
<div class="p">
<div class="note">
Note: In the following discussion we assume that the RSE server has been installed on Linux.
If you are running on a UNIX system the script name is "server.sh" rather than "server.pl".
</div>
If you do not have root access on a remote machine, you can start the server manually for your particular user id only.
Run the following commands:
<pre>
cd /opt/rseserver
perl ./server.pl [port]
</pre>
These commands run the server.pl script located in the /opt/rseserver directory. The port parameter to the server.pl script is
optional. If you do not specify a port, then the server will pick the first
one available and print the port number to standard out. By default, it is
usually 4033. If you would like to use a different port, you will then have
to enter this port number in port property for the Files subsystem for your
connection in the Remote System Explorer (see <span class="uicontrol">Connecting to the
Remote Server</span>, below). Otherwise, you do not need to change this
property.</div>
<div class="p"><div class="note">
Note: When you connect RSE to the server, the server will terminate as
soon as you disconnect the client. The daemon, however, will not terminate.
</div></div>
<p><b>Rexec Server Launcher</b></p>
<div class="p">If you have Rexec access enabled to your remote system, you
can also have the server started automatically by an Rexec command from the
client, when you connect. To do so, use the Server Launcher Properties in
the New Connection Wizard.
</div>
<p><b>Running the daemon at startup</b></p>
<div class="p">You might instead want
to configure the daemon to run at start up for Linux. To do so, you need to
append a call to the daemon to your startup script. Add the following lines
to the bottom of the /etc/rc.d/rc.local file:<pre>cd /opt/rseserver
perl ./daemon.pl &amp;</pre>
</div>
<p><b>SSL Encryption and Firewalls</b></p>
<div class="p">By default the RSE DStore connection is unencrypted. You
can, however, configure it to use SSL encryption.
<p>Because all dstore data transfer is done through a single TCP/IP
connection, the connection can also be tunneled through an ssh channel.
In fact, the same ssh channel can also be used to start the server,
like in the following example:</p>
<pre>
ssh -l moberhuber build.eclipse.org -L27127:build.eclipse.org:27127 \
"sh -c 'cd ~/rseserver/latest; perl ./server.pl 27127'"
</pre>
<p>
Here, the RSE Server is started on port 27127 through an ssh connection,
and at the same time port 27127 is forwarded through ssh to the local
host. You can now connect RSE to localhost:27127, and the connection
will transparently be forwarded to the remote system.
</p>
</div>
</div>
</div><br></br>
<p><b>Connecting to the Remote Server</b></p>
<div class="p">To make a connection
to your remote server:</div>
<ol>
<li><span>Switch to the Remote System Explorer perspective. From the workbench
menu, click <span class="menucascade"><span class="uicontrol">Window </span> &gt; <span class="uicontrol">Open Perspective </span> &gt; <span class="uicontrol">Remote System Explorer</span></span>.</span></li>
<li><span>In the Remote Systems view, <b>New Connection</b> is automatically
expanded to show the various remote systems you can connect to through the
Remote System Explorer. Expand <b>Linux</b> or <b>Unix</b> to
invoke the new connection dialog box and configure a connection.</span></li>
<li><span>Enter a name for your first profile and click <b>Next</b>. (This
step only occurs if you have never defined a connection before.)</span></li>
<li><span>Enter a connection name. This name displays in your tree view and
must be unique to the profile.</span></li>
<li><span>Enter the name or TCP/IP address of your Linux server in the <b>Host
name</b> field, for example, LINUX_A.</span></li>
<li><span>(Optional) Enter a <b>Description</b>. The description appears
in the Properties view after the connection is created.</span></li>
<li><span>Click <b>Finish</b> to define your system. </span></li>
</ol>
<div class="skipspace">
<div class="attention"><span class="attentiontitle">Attention: </span>To check your port number, right-click your
connection or subsystem from the Remote Systems view and select <b>Properties</b>.
Click <b>Subsystem</b> to view the relevant information. If your port is "0,"
then your Remote System Explorer communications server will pick any free
port on the server. If you specified a port number when starting the server,
you need to enter it here, for example, to work with a firewall.</div>
</div>
<div class="p">
<div class="note"></div>
</div>
</div>
<div>
<p><b class="reltaskshd">Related tasks</b>
<a href="tbeginwindows.html" title="">Connecting to a remote Windows
server</a>
<!-- TODO SSL configuration -->
</p>
</div>
</body>
</html>