Connections HOWTOs

This is the Connections section for the HOWTOs collection.

Connecting to a DAV Service

What is DAV?

DAV (pronounced as in davenport) is a way to provide file transfers via HTTP. In many respects, it can substitute for traditional FTP servers but provides a more robust, flexible, and secure way of sharing files.

Furthermore, most recent operating systems have built-in functionality that allows you to "mount" a DAV share just as you would a local network volume, so you can simply drag & drop files as well as access them directly without necessarily "downloading" them first or using a separate program.

For more information on DAV, please visit webdav.org.

Connecting

MacOS :: Windows 2000/XP :: Windows NT/95/98/ME :: Others

One small disadvantage of DAV-based file sharing is that the details of the connection method varies from operating system to operating system. Please see the particulars for your operating system of interest from the list above.

Another inconvenience concerns the use of "self-signed" security certificates. DAV clients will generally issue a warning that our certificate isn't from a "trusted" authority and require you to accept it despite this. Most clients also provide a means to accept our certificate permanently, in which case you will only have to do this once. However, the Windows certificate managment interface does not appear to do this properly and you will have to accept the warning each time you connect (and possibly multiple times per connection). Unfortunately, we know of no solution at this time.

Please have the following information (which should've been provided to you) before continuing: the URL you will use to connect to the DAV share, the login name, and the password.

MacOS

MacOS prior to MacOS X 10.4 (Tiger) do not have built-in support for SSL-encrypted DAV access. Please download and install Goliath and read its documentation.

For MacOS X 10.4 and higher, simply select Connect to server from the Finder's Go menu and enter DAV URL.

Windows 2000/XP

Windows 2000 and XP have built-in DAV support. For Windows 2000, make sure that you have upgraded to at least Service Pack 3- prior versions had known bugs. To make a connection, open the My Network Places icon on your desktop and then double-click the Add Network Place option.

On some installations/configurations of XP, you may not see My Network Places on the desktop- instead, please check your Start menu. You will have to select Next to proceed past the opening description, followed by Choose another network location when asked where your network place is.

In the subsequent dialog, enter the URL you were given and press Next. At this point you will probably be given a certificate warning- accept it. You should then be prompted with a login dialog - enter your login name and password and then Finish the wizard.

Your My Network Places window should now have an icon for the DAV share and it should open if your connection was successful. That icon should persist so for future connections, simply open it and login as usual.

You may be repeatedly asked to accept a certficate warning, or asked to login again despite having already done so. Just repeat as necessary. For details on how you can permanently avoid these warnings, see the HOWTO on installing 3rd-party certificates.

Windows NT/95/98/ME

The base Windows NT/95 operating systems did not natively support DAV. However, depending on the service pack level, Internet Explorer version, and/or Microsoft Office version, DAV support may have been added to your computer. If, when you open My Computer, you see a folder named Web Folders ("Web Folders" is what Microsoft calls DAV shares), this capability has been installed.

Windows 98/ME should have this capability installed by default.

To connect, open the Web Folders folder, then launch the Add New wizard. Enter the URL when prompted, accept any security certificate warnings, and then login with your provided login name and password. An icon for the DAV share should appear in that folder and persist for future connections.

Other operating systems

Please consult the webdav.org projects page for a listing of clients for a variety of platforms.

As a fallback option, it is also possible to use a standard web browser to access a DAV share, since it runs over HTTP. Just point your browser at the proper URL and login as usual. This method, however, will not let you download more than one file at a time, and provides no way to upload files, so is not recommended.

How To Install Microsoft UAMs for Macintosh

Step 1: Open the network browser

In the Apple Menu, click on the Network Browser.

Step 2: Locate HGSERV2 on the AppleTalk Network

After the Network Browser has loaded, look for the AppleTalk icon. Click on the arrow to the left of the AppleTalk icon to expand the AppleTalk Network.

Next, click on the arrow on the left side of the Gonda_HG icon to expand the Gonda_HG network

Now, locate the HGSERV2 icon and double-click on it.

Step 3:Choose Your Log On Method

A window may appear that will ask you which type of "log on method" you would like to use. If Microsoft Authentication is displayed as an option (or Microsoft UAM), then Microsoft UAMs are installed, and you need not continue. Otherwise, click on Apple Standard UAMs, and then click OK.

In the next window that appears, select Guest so you don't need to type in any password.

Step 4: Open the Microsoft UAM Volume Folder

A list of names (like in the picture shown below) will appear after you have successfully logged into HGSERV2. Find the icon labelled Microsoft UAM Volume and double-click on it.

Step 5: Locate Your AppleShare Folder

Set the open window from the last step aside for now--we will need it later. Now open the AppleShare folder on your computer. To do this, first double click on the Macintosh HD icon on your desktop. This is your hard drive.

In the window that appears, find the System Folder and double-click on it.

Step 6: Copy the AppleShare Folder

Now, you have to copy the AppleShare folder from the network to your computer--that is, you will copy the AppleShare folder found in Step 4 to the system folder found in Step 5. If you receive a warning when attempting to copy, read below. If you receive no warning messages, proceed to Step 7.

You might receive a warning (like the one below) when trying to copy. If so, click Cancel. Otherwise, if you do not receive this warning, please continue to Step 7.

To get around this warning, double-click on the AppleShare Folder in the System Folder window from Step 5.

In the Microsoft UAM Volume window from Step 4, double-click on the AppleShare Folder.

Copy the MS UAM icon to the AppleShare Folder on your hard drive.

Step 7: Reboot Your Computer

Congratulations! You're done installing Microsoft UAMs. All that is needed now is for you to restart your computer. You are now ready to use Microsoft UAMs!

How to Set Up Your Personal Drive on a Macintosh

General Information

Before you set up your personal drive, please e-mail computersupport@mednet.ucla.edu, or stop by our office to inform the Computer Support team. A quick initial setup must be completed before you will be able to access your personal drive from a Macintosh.

We highly recommend that you use Microsoft UAMs (User Authentication Modules) when connecting to your personal drive from a Macintosh. This will insure the highest level of security when transmitting your password over the Gonda network. If you do not have this installed, please visit the How To Install Microsoft UAMs for Macintosh page.

Step 1: Open the network browser

In the Apple Menu, click on the Network Browser.



Step 2: Locate HGSERV2 on the AppleTalk Network

After the Network Browser has loaded, look for the AppleTalk icon. Click on the arrow to the left of the AppleTalk icon to expand the AppleTalk Network.



Next, click on the arrow on the left side of the Gonda_HG icon to expand the Gonda_HG network



Now, locate the HGSERV2 icon and double-click on it.



Step 3:Choose Your Log On Method

A window will appear that will ask you which type of log on method you would like to use. We highly recommend using Microsoft Authentication (also called Microsoft UAMs). If you do not see an option for Microsoft Authentication, click here for a guide on how to install Microsoft UAMs.

Continue to Step 4a after the Microsoft UAM has been installed. If you insist in using Apple Authentication (which transfers your password in unencrypted, continue to Step 4b.

Step 4a: Logging in with Microsoft Authentication

Click on Microsoft Authentication and then click OK.



In the next window that appears, enter your Gonda user name and password and click OK. The Gonda user name and password is the name and password you would type if you were logging into a Windows computer on the Gonda network.



Step 4b: Logging in Using Apple Standard UAMs

Click on Apple Standard UAMs and then click OK.



In the next window that appears, enter your Gonda user name and password and click OK. The Gonda user name and password is the name and password you would type if you were logging into a Windows computer on the Gonda network.



Step 5: Create a Desktop Shortcut

A list of names (like in the picture shown below) will appear after you have successfully logged into HGSERV2. Find your username, then drag and drop it onto your desktop. This will create a convienent shortcut from which you can access your personal drive.



You're Done!:

Congratulations! You can now use the icon you've just created to access your personal drive whenever you want. Follow this same procedure on another Macintosh to access your user space.

How to access Gonda servers remotely?

Titan titan.genetics.ucla.edu is our Windows 2003 based terminal server to enable remote access to Gonda servers. To determine if Titan is right for you, please read on. If you know for sure Titan is what you need, you may jump to this setup how-to - Setting up Remote Desktop Connection for Titan

Where to use Titan

Titan is useful for remote connections within or away from Gonda, such as:
- Work from home
- Work from another office/building
- When your office computer doesn't have the application that Titan has installed
So your use isn't just limited to being "away from Gonda"

Advantage

Titan is useful for those who work from home but need files or applications on the Gonda servers-
1. Accessing your Mednet email personal folder (.pst) while not at Gonda.
2. Accessing your Windows home directory (M drive) etc while not at Gonda.
3. Moving files from your M drive etc onto local machine (i.e. moving a file saved on the Gonda file server to C drive of your home computer)
4. Generating .pdf files (server has the Adobe writer installed)
5. Running programs that are available at Gonda but not available elsewhere. (i.e. Licensed programs that are not installed at home or lab computer but a license is purchased for Titan i.e. Photoshop, Endnote etc) Check out our list of software installed on Titan below.

Disadvantage

Titan mimics your access in front of a computer and your experience will depend on the speed of your computer and Internet connection. The following are things that would affect you adversely.
1. If you have a dial-up connection, you will notice extended latency; graphics being refreshed slower.
2. If you have an unreliable connection, your connection to Titan may also be interrupted.
3. You will also notice extended latency working with programs that are graphically intensive.
4. If the server is extremely busy, your application's performance will also be taxed, as the server is being shared by other remote users. However, this server is powerful enough for our current demand that you should not notice a significant change.

Setting up Titan

If you're ready to set it up on your computer, please go to: Setting up Remote Desktop Connection for Titan.

Sample List of Software on Titan


-7-Zip
-ActivePerl 5.8.6
-Adobe Acrobat 7
-Adobe Photoshop Elements 3.0
-Adobe Reader 8
-AxioVision LE
-Citrix Program Neighborhood
-Compaq Fortran 6.5
-D-7000HSM
-Endnote 5
-Firefox
-Financial Projection Module (FPM)
-Genoseq ODBC setup
-Ghostview, Ghostscript
-GoDiagram Win 2.2
-Gregor
-HostExplorer (OASIS, AIS access)
-HP 3500 laserjet printer driver only
-Intel Fortran Compiler 8.1, 9.0, 10.0, 10.1
-Linkbase, with PDOX file
-MendelPro
-MikTex 2.7
-MicroSpell
-MS Office 2003 Suite
-MS Visual Studio .NET
-MS Visual Studio .NET 2003
-MS Visual Studio 2005
-MW Snap 3
-MyGeneration
-NDoc 1.3
-NX Client for Windows (nomachines)
-PCTex 5
-PDFCreator
-PDFill
-Progeny 6
-Python 2.5, 2.6
-QDB files (c:\qdb)
-QDB ODBC setup
-Quicktime
-R 2.1.1
-Sequencher Demo
-Shockwave player
-SPSS 12.0
-SQL Enterprise Manager
-SSH Client 3.2
-TeXnic Center
-TortoiseCVS
-Wavemaker 4.1
-WholePathwayScope
-Wink
-WinRAR
-WinSCP3
-WinZip
-Xwin 32
- More upon request

Mapping Network Shares Permanently on a Windows Client

To map a network share permanently on you computer, right click on the "My Computer" icon and select "Map Network Drive" from the drop down menu.


Now you need to select the letter that will be used to represent this share. You want to pick a letter that is not already in use. A safe bet would be to pick letters towards the end of the alphabet. Make sure to check "Reconnect at logon" so that you never have to map the drive again on this system.



Now, click finish and you are done

NX connections to Calypso

What is NX/Nomachines?

NX (or nomachines) is a type of connection method that allows a graphical login to calypso, our Linux-based cluster head node. Calypso has NX server software on it and you use a special NX client on your workstation (available from nomachine.com. The end result is something similar to X11-over-ssh but easier to use and much quicker. It's more like using remote desktop to Titan.

This document covers how to configure the NX client as a user of calypso. It does not cover how to install NX client- please contact us for assistance with installing the software.

You will also need a calypso key file - please contact us for this when you have the client installed. It should be named calypso.key.

Configuration

Configuration needs to be done once- once you have configured it, you will not need to do so again for your user account on the particular machine you configure.

To run NX client for the first time, look for it in your Start->Programs menu. You will be greeted by this wizard screen. It's not very informative, but Windows applications seem to like to reassure users about things. Just click the Next button, as everyone has been trained to do by such things.

The next screen asks you what the session parameters are. The session name is just a name you will use for this series of settings - just put calypso. The hostname is a bit more important: is should be exactly calypso.genetics.ucla.edu. The internet connection setting can be left at the default if you are configuring within the department. If you are setting up a system to connect to calypso from home, you may want to drop it down to IDSL or Modem.

Despite what this screen may lead you to believe, you're not really done yet. Click the Finish and ponder the inappropriateness of that label.

This is the login screen, but you are not ready yet to login. Instead, you need to do some more configuration, so click the Configure button.

OK, so things look a lot more complicated now. So much for the wizard. The first thing you might want to do is change the desktop type next to Unix to GNOME from the default (which should be KDE. You don't have to, but GNOME is something more users are familiar with.

Next, click on the Key button to configure the login key (which you should have on hand somewhere).

Here, you need to click the Import button and navigate to the key file you were given (ie calypso.key). Once you have done so, and loaded it, click the Save.

OK, now you're done. Click OK until you are back to the login screen and enter your calypso username and password before clicking the Login button. After a short wait, you should be logged into calypso's graphical environment.

If it's the first time connecting and you get an error, please try quitting the NX client and restarting it (sometimes it needs a little shaking). If you continue to have trouble, please contact us via the helpdesk and provide the text of the error log that appears under Details.

Secure X11 Connections

You can make a graphical connection to calypso using what is called "X11". With it, you can run programs that have a graphical interface (such as Gregor for Unix) and manage your files via a more friendly graphical interface than the standard Unix command line.

However, the setup of such a connection is a little bit more involved that your standard ssh connection.

Configuring X-Win


First, you need a program called X-Win32, which provides the actual graphical display. If you don't already have it installed, please see contact the helpdesk for assistance.

Once it's been installed, first run the configuration program, usually found in Start->Programs->X-Win32->X-Config. Click on the Add button to configure a session.



In the subsequent dialog, you'll be asked what connection type you would like to use. Select the XDMCP option, and then click the Next button.


In the next dialog, enter calypso for the Session name, select Broadcast for the XDMCP type, set the Window mode to Single, and check the Launch New Window option. Once you're done, hit the Save button.


Now, go Window tab, and select Single for Window Mode. Click OK when you are done to complete the X-Win32 configuration.

Now, you can run X-Win32. Nothing will happen yet- the X-Win32 icon will appear in your system tray and you may see a blank gray screen that you can hide for now.

Connecting and tunneling with ssh

To actually make the connection to calypso, you will need to ssh to it first. For this, you will need "SSH Secure Shell". This is installed on most computers in our department. If you can't find it under your Programs menu, then contact the helpdesk for assistance.


If you haven't already configured a "profile" for calypso, select Add Profile..." from the Profiles drop down menu, then give it a name (ie "calypso"). Once the profile has been created, you can go to the Edit Profiles option.


The profile editor may look different, depending on your version of SSH Secure Shell. Make sure you select the "calypso" profile you created above. The important configuration option is the hostname, which should be "calypso.genetics.ucla.edu". The username is convenient but not necessary.


The other important option is called "X11 Tunneling", and in this version of SSH Secure Shell appears in the Tunneling tab. Check the box to turn it on.


Once you have set these options (hostname and X11 tunneling), you can save by clicking OK. Now you are ready to connect. Select calypso from the Profiles drop down menu and you'll be asked for your calypso password. Once you've succesfully logged in, enter the following command to begin your X11 session:

gnome-session

Once your session has begun, you can hide the SSH Secure Shell window.

Setting up Remote Desktop Connection for Titan

There are 2 parts to connecting to Titan:
a. (one-time) Request to have Gonda account enabled for Titan access, and get "pcf" file
b. (recurring) login setup


a. Request to have Gonda account enabled

****Before logging onto Titan for the first time, please enable your Gonda account with Titan access with the IT office via the Helpdesk **** you'll also receive a "pcf" file that's needed for the setup.

b. Login Setup

If you have Windows XP or newer, you may skip ahead to step 2 of Login Setup.

If you're running Windows 2000 or older: You need to first download the Remote Desktop Connection client from Microsoft for free.

Windows: Windows XP Remote Desktop Connection software

Macintosh: Please read the system requirements before proceeding Microsoft RDC Client for Mac 2.0.1

b. Login Setup - step 1

After downloading, double click on the file (with file extension .exe for Windows, .dmg for Mac) to begin installation.




On the PC, after the installation, you can find the program under Programs; Accessories; Communications or alternatively Programs; Accessories; Remote Desktop Connection.

On the MAC, copy the Remote Desktop Connection folder onto your Macintosh HD. Afterwhich you may move the .dmg file along with the expanded folder into the trash, leaving only the copy on the HD active.




c. Login Setup - step 2

The following steps apply to both PC & MAC--

Opening the installed program (Remote Desktop Connection) yields this dialog box where you should type in titan.genetics.ucla.edu:33389

REPEAT ONLY THIS STEP for future logins.

If you like to set up Outlook, please try this link: Outlook Setup

Optional Fine Tuning

If you like to fine tune your connection, you may perform the following steps-




DISPLAY: Tune the desktop size to best reflect your local monitor size and/or how you like the terminal screen to appear. You may set it to full screen to fully mimic working at the server, or set it the size of a typical window so you can go back and forth on the server and local desktop.




LOCAL RESOURCES: If you like to move files back and forth between the server (i.e. M drive) and your own computer, you may want to map your drives. By clicking on "Disk Drives", all of your own computer's disk drives will be listed as additional network drives in the Windows Explorer on Titan, denoted as "C on COMPUTERNAME" etc.




EXPERIENCE: You may also reduce bandwidth by minimizing unnecessary traffic by unchecking all the boxes in the Experience tab.

Setup secure X11 connections using cygwin

Thanks to: http://x.cygwin.com/docs/ug/setup-cygwin-x-installing.html (that URL includes a few pictures, in case something is unclear)

GOAL To setup cygwin on a Windows workstation so that you can forward X11 apps from a Unix workstation.

ALTERNATIVES There are no other free and opensource alternatives to cygwin for X11 forwarding to Windows workstations. You can use commercial products such as Hummingbird eXceed, X-Win, or NX.

INSTALL cygwin

• goto http://cygwin.com/
• click on Install or update now! and pick a place to download setup.exe
• after downloading, run setup.exe
• choose to Install from Internet
• all other defaults are fine
• when asked for mirror, you can choose any, but http://mirrors.kernel.org/ is a pretty fast one
• when asked for which packages, there are two required to make this work:

      Net > openssh (which auto-selects openssl)
      X11 > xorg-x11-base (which auto-selects other X11 packages)
  

  ... feel free to include anything else you want

Now you've finished downloading and installing cygwin.

USE cygwin

• double-click on cygwin icon (this opens up a terminal window, where you have a more Unix-like command-line environment within Windows)
• to do X11 forwarding, type startxwin.bat (this will open up another terminal window with all the X display information already configured by cygwin)
• in this new terminal window, type
• ssh -Y ip.address.of.unixserver
• login with your credentials
• now you can run any X program and they will be forwarded from the Unix workstation to your Windows workstation
• examples:

      gnome-session (for an entire desktop)
      thunderbird (a particular app)
  

• you can run the programs in the background, which gives you back control at the command-line, thus you can forward multiple X11 apps:

      thunderbird &
      gaim &
  

In subsequent sessions, you only need follow the steps under USE cygwin.

Transferring files with ssh.com Secure Shell

In addition to its primary function as an ssh terminal (ie command line) program, ssh.com's Secure Shell also provides built-in file transfer features that allow you to move files to and from our Unix-based servers. The following describes how to do so. However, please note that installation and configuration of the software is presumed and not covered here. If you haven't already installed and configured the software, please see our other HOWTO for details first.

There are two ways to begin a file transfer, depending on whether you are already connected to the server in question via terminal, or whether you are only doing a file transfer.

Starting a transfer during a terminal session

If you are already logged into the server, simply go to the Window menu and select New File Transfer.

A new window will appear with two panes. The left pane represents your local file system and the right pane represents the remote side (the server).




Starting a transfer outside of a terminal session

You can also transfer files without having to initiate a terminal connection at all by launching the the file transfer program directly. Typically, this is found under Start->Programs->SSH Secure Shell->Secure File Transfer Client.

Once you've launched the file transfer client, you will need to connect to the server. Go to the Profiles submenu and select the profile for the setting to which you wish to connect (see our HOWTO on setting up ssh.com Secure Shell for details).

Once you've connected, you will be presented with a two pane window as described above: the left pane is your local file system and the right pane is the server.




Transferring files

There are a few ways to transfer files. If you navigate the filesystems using the panes, you can right-click on a file and select Upload or Download from the context menu, depending on whether you are right-clicking in the local pane or the remote pane.

When using this method, please be mindful of the location of the target pane, otherwise, you may have difficulty finding your file.

Another way to transfer files is to drag-and-drop it. If you drag a file from the server pane to your Windows desktop or another folder, it'll transfer it to that location. Conversely, you can drag Windows files to the server pane to upload. The pane at the bottom of the window displays a list of the files you have transferred during your session. This is useful for checking on the status of an ongoing transfer and also for keeping track of what you have already transferred.

Using CuteFTP for FTP/SSL connections

In order to manage content on the Biostatistics website, you will need to use an FTP/SSL connection to the webserver in order to transfer files back and forth. For this, you will need an FTP client that has SSL (encryption) features - CuteFTP is one that is licensed by UCLA and available from the BOL Software page.

You will also need an account on the web server. Please contact the helpdesk if you need one.

Using CuteFTP

When you start CuteFTP, cancel any "Use Wizard" or "Setup Connection" dialogs. From the main screen, click on the downward pointing arrow to the right of the flower document as shown here:



From the menu that appears, select FTP/SSL Site

In the connection dialog box that appears, fill out all relevant info (label, host address, username, and password) as indicated here:



Once it has been filled out, click the Connect button. You'll then be presented with a view of the web server's contents in the right side pane. You can drag files back and forth between the left side (which represents your own computer) and the right side as appropriate.

On subsequent uses of CuteFTP, you can select this "profile" via the Connection Manager that opens at launch.

Using Putty for SSH connections

PuTTY is a free ssh client that can be used to connect to our Linux servers. You can read about PuTTY at http://www.chiark.greenend.org.uk/~sgtatham/putty/, or jump directly to the download page (there are many files listed - you only need putty.exe).

Once you've downloaded putty.exe, you are ready to run it. There is no installer, just the single file executable, so it can be placed anywhere you'd like (the desktop is a convenient place for most). Simple double-click on the putty.exe icon to get started.

Configuring a session

When you first launch PuTTY, you'll be presented with session configuration options. Here you can can create new sessions and select from existing ones. To create a new session, enter the hostname as indicated, the select SSH for the protocol.

To save this session, next enter a name for it in the Saved Sessions box, then click the adjoining Save button. Your session will now appear in the session list.

Finally, to actually start your session, double-click on the session in the list, or select it and click the Open button at the bottom of the window.

In subsequent uses, you will not have to enter the hostname and select the protocol, only the first time when you create the session.

The first time that you connect to a server, PuTTY will present a warning about the server host key - this is normal behavior. Simply click Yes to accept the key and you won't be presented with the warning when you connect to that server again.

Finally, you will be presented with the actually ssh session window and asked for your login/username and password.

A related HOWTO on using WinSCP to transfer files is also available.

Using Tortoise CVS for Revision Control

TortoiseCVS is a free CVS client for Windows- it speaks the CVS standard for revision control of files (typically program source code). It can work with a local "repository" (where your files and information about revisions are kept), or as described in this document, access files kept on a Unix-based server over an ssh connection.

If you plan to make frequent use of CVS against an ssh server, please make sure to investigated the advanced ssh instructions to avoid having to enter your passwords for every CVS transaction.

Getting Started

First, download TortoiseCVS at http://tortoisecvs.sourceforge.net/. Run the downloaded installer to install TortoiseCVS on your system (administrative privileges may be required).

TortoiseCVS doesn't function as a separate program, per se. Instead, it acts primarily as an extension to the Windows file manager and provides context-sensitive actions that can be accessed via right-click access. For example, if you right-click on the desktop after installing TortoiseCVS, you should see two new entries: one called CVS Checkout and one called submenu called CVS. Preferences for TortoiseCVS can be accessed from this CVS submenu, but the defaults work fine for most purposes.

Creating a CVS repository

Before you begin using CVS to manage your files, you need to know where your repository will reside and initialize it for CVS use. This can be on your local computer, or in our case, somewhere on the Linux server. A convenient place is a directory named "CVS" in your home directory. Thus, first log into your Linux account and create a directory named CVS by executing CVSROOT=/home/jbruin/CVS cvs init (users of csh/tcsh will have to alter this syntax).

The above only needs to be done once for the repository and each repository can contain as many projects as you desire, so typically, it will only have to be done once ever.

Defining and new module

You are now ready to import files into CVS. Typically, these files will reside in a directory on your Windows workstation and in CVS parlance will be called a module. Right-click on this directory and under the CVS submenu, select Make New Module...

You'll then be presented with a complicated dialog requesting details for where and how to store your files. Fill in as follows:

• Protocol: select Secure shell (:ext:)
• Server: enter calypso.genetics.ucla.edu, or the appropriate address of the Linux server
• Repository folder: enter /home/jbruin/CVS, or the appropriate full path to your repository.
• User name: enter your username for that server
• myprogram: enter a name for this module. Typically, this would be the name of your program. Generally, it is advisable to exclude spaces and punctuation marks in your module name.

Once everything is set, click the OK button. Note that you need to enter this information only once per module; subsequently, you can select the module details from the menu that appears at the top of this window.

If everything has been entered correctly, you should see a log window appear and then a password dialog. Enter your password (it may request it more than once). Eventually, you should see a success message - your new module has been created.

Adding files to your module

Next, you need to import all your files into the module (the step above just prepared the module). Right-click on the directory again and this time select CVS Add Content.... You may notice that the icon for this directory now appears different, probably with a green mesh overlay - this is the indicator that this directory is under CVS control. Once you've selected CVS Add Content..., you be presented with a dialog listing all of the files in that directory. Go through this list carefully and uncheck any files that do not need to be revision controlled. Typically, only source files need revision control - compiled binaries, temp files, object files, almost all binary format files shouldn't be placed in CVS. Take care to do this during the importing step as removing files later is troublesome and will continue to take up space in your repository (since it is a permanent history of your module's files).

Once you've properly marked the files you want to import, click OK. Again, you will be asked for your password, possibly many times (once each for each subdirectory). Enter it as often as it asks. Eventually, you should see a success message.

Commiting your first version

Now that CVS knows about your files, you must commit. Commiting is CVS's way of "saving" changes and actions, so until you perform a commit, nothing has been "saved". In our case, we've told CVS that we have a bunch of files for this module, and now we have to tell it to make those additions permanent. To do so, again right-click on the directory with your files and this time select CVS Commit.... This will look similar to the file addition dialog above, but there will be an additional box where you should type in a comment describing what you are committing. Here, we are simply importing all files, so the comment can be "Importing". When you later subsequently commit changes to files, you should describe in more detail what changed between commits.

Basic Usage

After making the initial commit, your project and all the specified files are now under CVS management. For the most part, this is not something you have to give much thought to as you work with your files. However, there are two actions you should keep in mind: updates and commits.

A CVS update attempts to sync your local directory with that in the repository and report any changes. Typically, you will see 3 classes of files when you do a update:

1. Files that match between your local copy and the repository (ie no changes have occurred)
2. Files that you have changed but not yet commited - these will appear with a flag M and will not be uploaded to the repository until you do a commit
3. Files that have changed in the repository but your local copy is out of date. This will occur, for example, if two or more developers are working with the same repository and another developer has commited some changes since the last time you did an update. In this case, these files (marked U) will be downloaded to your local copy of the module. This will also occur if you are missing a file from the repository (for example if you deleted one by mistake, or if another developer added one).

Another, less common, occurrence is a combination: you have modified your local copy of a file and another developer has committed changes. In this case, CVS will try to reconcile both sets of differences and will warn you of the circumstances. Typically, it will mark areas of the file in question where conflict arises and you must reconcile them in some way.

One other action that will frequently arise is the addition of a file. Simply creating a new file in your local copy of the module is not sufficient to make it part of CVS management: you must add it (right-click and select CVS Add) and then commit it as usual. Conversely, don't just delete a file from your local copy to remove it- when you next execute an update, it will re-appear. Instead, right-click and select Remove from the CVS submenu.

Advanced SSH configuration

During large updates and commits, you may find yourself having to enter your password quite often, depending on how the number of subdirectories in your CVS module. With a little work, you can set up ssh to only require you enter your password once per session by using ssh's key-based authentication.

First, you will need a couple of additional ssh tools. Visit the PuTTY home page and download the following 3 programs:

• PuTTY
• Pageant
• PuTTYgen

Creating your keys

Place these 3 executables in a directory somewhere (for example C:\putty). Run puttygen first. Select the SSH2 RSA option at the bottom of the screen, then click the Generate button. This will take a while, and you'll be asked to move your mouse around. Once this is done, enter a passphrase twice, where indicated. This should be different from your password, and should be a sentence. For example My username on calypso is jbruin (Do not use this, or a variant, as your passphrase!).

Now save your key by clicking the Save private key button and storing this in a safe place (for example, in My Documents). Remember where you saved this file. If you lose it, you'll have to go through this again.

Next, copy the contents of the text box near the top, the one that looks like gibberish, into your clipboard. You will need to create a file in your Unix home directory at ~/.ssh/authorized_keys2 with this as its contents. Make sure that the directory and file have restricted permissions by executing chmod 700 ~/.ssh and chmod 600 ~/.ssh/authorized_keys2.

Finally, configure putty to mimic the TortoiseCVS connection to the CVS server. Run putty and in the configuration, select SSH for the protocol, enter your hostname (ie calypso.genetics.ucla.edu), enter a session name, and save your settings.

Using your keys

Now, run pageant, which is the authentication agent that will manage your keys. After double-clicking on pageant, it will appear in your Windows system try (usually on the right side of your task bar). Right click on it and select Add Key. You'll then be asked to locate your private key file and enter your passphrase.

If everything has been configured correctly, you should be able to perform CVS actions without having to enter your password each time. Instead, pageant will manage the connections and provide the authentication credentials for the keys you created.

Using WinSCP for File Transfer

What is SCP? Why should I use it?

scp is a file transfer protocol not unlike FTP - it is a means to transfer files between client machines and Unix-based servers.

The difference is that scp is secure. The ftp protocol uses cleartext passwords than can be sniffed/intercept by malicious individuals. These passwords can then be used to compromise your account. scp, on the other hand, encrypts all data transfers so passwords cannot be intercepted.

Obtaining WinSCP

Using scp is quite easy, but as it is a separate communications protocol from ftp, you will need to obtain a new client program that can speak it. WinSCP is one such program that is as easy to use as most ftp programs and is free. It can be obtained from http://winscp.sourceforge.net/eng/. Look for the download link for the exe version, and save it to your desktop or where you prefer to install programs.

Using WinSCP


When you first run WinSCP, you will be presented with the main login configuration screen. Here, you will need to enter the address of the server you wish to conect to and your username.

If you are using a computer that only you use, then you might want to enter a password as well. If you are using a shared computer, it would probably be best not to enter a password at this time (you will be asked later).

Once you have configured the host name and user fields, click the SSH tab at the left sidebar of the window.





In this tab, make one change to the WinSCP default settings: for Preferred SSH protocol version, select 2

If this is the first time connecting to this server, click the Stored sessions tab next so that you can save your settings.





Here you can save your settings so that you won't go through the above each time. Click the Save button and give the session an appropriate name (usually the default is OK).

When you run WinSCP again, you will be presented with this screen first, so you can simply select the particular settings you want to use for login. If you wish to configure a connection to a new server, select the Basic tab and repeat the steps above.





Once you have all your settings configure, you can click the Login button at the bottom of the window.

The first time you connect to a particular server, you will see a notice about host keys and trust. You can click the Yes button - you will only see this the first time you connect to a server.

If you didn't configure a password above, then you should be presented with a password prompt.





Once you have sucessfully logged in, you will be presented with a two-pane window that should be familiar. On the left is a representation of your local (Windows) computer's file system. On the right is the remote (Unix) file system.

To transfer files first make sure to browse to the proper destination in the appropriate pane. Then browse to the file(s) that you want to transfer in the source pane (you can use control-click to select multiple files). Then press the F5 Copy button at the bottom of the window. Alternatively, you can press the F5 key or drag-and-drop files to initiate transfers.

Using ssh.com Secure Shell for SSH Connection

Obtaining the SSH client:
You can obtain SSH by requesting it from the IT Office or you can download SSH for your PC or Unix system at these locations:
http://charlotte.at.northwestern.edu/bef/SSHdist.html
http://www.sfsu.edu/~helpdesk/ssh/ssh329/
http://www.filewatcher.com/m/SSHSecureShellClient-3.2.9.exe.5517312.0.0.html



Installing SSH:

Double Click on the .exe file for the SSH Client

Note:
Version 3.2.9 is the one that we will be covering in this Howto. This is the last version of the Secure Shell software from ssh.com that is free to download and use. If you would like to purchase a newer version, you can go to http://www.ssh.com

After you double click the executable, the installation process will start. Go through the steps of the installation and select your desired choices. Soon, you will see a window indicating the installation has been completed.




Configuring SSH:

The last step is to configure SSH.

Run SSH, then go to profiles and click on Add Profiles




Next, type in the name you want to use for your new profile and click on the Add to Profiles button.

Now, go to Profiles again and select Edit profiles this time. From the new window, select the profile name you just created. Make sure to put in the host/server name (shown here for calypso.genetics.ucla.edu) and your user name. All other parameters should be set to to what you see here. Then click OK.


Now go to Profiles yet again and select the server you just configured (calypso). Upon your first connection attempt, a window like the one below might pop up and ask you if you want to save the host key. Click Yes.

You should then be prompted to enter in your password in a window like the one below. Put in your password and click Ok.

If your password was correct, you will be brought to the command line of the shell connection. You are done!