Tutorial - Clients

From distributed.net
Jump to navigationJump to search

Introduction

First, we'd like to thank you for for taking an interest in distributed.net and the distributed computing projects we undertake. This brief introduction to the client software will get you started in participating in these projects. Please note that you are only permitted to run the client on computers you own or have permission from the owners to run the client. For more on this important topic, please read thoroughly our Official Policy page.

Downloading the client

If you have not yet obtained a client, you can download the latest client from the clients page. Start the download by clicking on the link for your platform.

Alternatively, the clients can be downloaded via anonymous ftp from the ftp site at ftp://ftp.distributed.net/pub/dcti/current-client/.

If you are using a computer running Windows or MacOS, the software will be normally automatically decompressed. If this doesn't happen on your system you will need to download WinZip for machines running Windows or StuffIt Expander for MacOS.

For Unix and MacOS X clients the tar gzip archive needs to be expanded and untarred. This can be done using the command "tar -xzf dnetc...tar.gz" or "gunzip -c dnetc...tar.gz |tar xf -" for systems without GNU tar such as Solaris and HPUX. The client should be downloaded to a user account, not root.

If you don't find a client for your platform, check our bugzilla port requests page and if the port for your platform has not been requested yet, you can request one.

Once you have the client on your system, you need to configure it.

Configuring the client

If you have never run the client before, it will initiate the menu-driven configuration. Save and quit when done, the configuration file will be saved in the same directory as the client. Then, simply restart the client. From that point on it will use the saved configuration.

The configuration options are fairly self-explanatory and can be run at any time by starting the client with the '-config' option. A list of command line options is available by starting the client with '-help'.

Many of the configuration options can be left at the default values, but it is important that you enter your e-mail address so that the work your client does in attributed to you in the project statistics.

Although most people's connection to the Internet involves no special client settings, addition configuration is often required for systems behind firewalls and there are also systems not connected to the Internet which can be used provided they are updated by means of floppy disks. Additional information for such cases is provided in the Network Setup Guide.

Users of computers running Windows have two installation options described below.

Windows - Service

A Win9x/WinNT service is a process that starts running when the computer is powered up, and does not stop until the computer is powered down. Unlike normal processes, services will continue to run even when a user logs out.

Obviously, a service does not have access to remote file and disk resources that require a user to log in first. (this restriction can be nullified on NT - see FAQ in the client Readme file).

To install the client as a service, run the client with the -install switch. This will cause it to register itself with the service handler, which will in turn automatically start the client the next time the computer is powered up. To deinstall a client previously installed as a service, start it with the -uninstall switch.

Windows - Screensaver multiplexor

Beginning with v2.7112.444, the client release archive includes a screensaver multiplexor.

Note that the screen saver and the client are different executables. The screen saver multiplexor (*.scr) is of little use without a client (*.exe).

Under normal circumstances, having a client running continuously in the background is usually preferred since it continues to use the idle processor cycles even when a machine is actively in use. For instance, when typing a letter, the machine is really mostly idle, waiting for keystrokes, and the client can use that time.

However, in some environments, utilizing *any* cpu time while a user is actively trying to use the machine may be a problem. The distributed.net screen saver multiplexor addresses that issue. When activated by the system, it will launch a client in the background and _another_ screen saver of your choice in the foreground. If a client is already found to be running when the screen saver is activated by the system, it will not be started a second time.

Not only does this permit an installation where use of any cpu time is a problem, it also resolves two other issues that a client-as-a-screensaver cannot: a) it does not requires storage of client buffers and/or configuration file in the windows (or windows/system) directory and b) it allows a user to continue to use his/her preferred screensaver.

Installation involves...

  1. Simply copying the *.scr file included with the client distribution to one of the following locations:
on Win95: your \WINDOWS\ or \WINDOWS\SYSTEM\ directory
on WinNT: your \WINNT\ or \WINNT\SYSTEM32\ directory
  1. Once you have copied it, go to the Display properties in the Control Panel and select the distributed.net screen saver as the active screen saver.
  2. Clicking on the "Settings..." button opens the multiplexors configuration dialog. Select the client you wish the multiplexor to run in the background and select the screensaver you wish to run in the foreground.
  3. If the client had been previously configured, there is nothing else you need do. Otherwise click on the "Configure" button to configure the client.

Note for Windows NT 3.x: The screensaver API changed between NT 3.x and NT 4.x to support the "mini-preview" mode first introduced in Windows95. The multiplexor supports the new interface, but since a 4.x system will not start a 3.x screensaver "correctly", the multiplexor was "branded" as a 4.0 executable. Beginning with 2.8001, the _client_ will watch for this condition on startup, and if it finds itself in a 3.x environment, will re-brand the .scr as a 3.x executable.

Running the client

Windows

If you have installed the client as a service, you can start the client in Windows 95/98 using "run dnetc.exe -hide", in NT "net start dnetc", or restart the machine which will start the client in both 95/98 and NT. The client will automatically start with the screensaver if installed as Screensaver Multiplexor.

MacOS

Simply double click the dnetc application icon to start the client running and use File/Quit or cmd-Q to quit the client. The MacOS X client is started from the command line the same way as Unix clients.

Unix

If you are using Unix, you might want to use "screen" (available from ftp://ftp.gnu.org/pub/gnu/screen/) to run your client so you can check on it, or you can choose to run it quietly as a daemon by typing:

./dnetc -quiet

Detailed platform-specific installation, configuration and operation instructions for clients for other platforms can be found in the client Readme files.

More information

You can get more information using distributed.net FAQ-O-Matic and help facilites described in the distributed.net Help Facilities page.