CST.net Technical Questions

More Advanced Questions

What is a protocol?
For our purposes, a protocol is an agreed upon method of communicating information between two computer systems.
Back to top

What is a connection oriented protocol?
A connection oriented protocol is a protocol that maintains at least one persistent connection between the client and the server. An example is FTP, during an FTP session the client and the server maintain 2 continuous, persistent connections with each other, a control connection and a data connection. Connection oriented protocols are 'bandwidth hogs' because the client and server must constantly be in communication with each other, regardless of whether or not any useful data is being transferred.
Back to top

What is a connectionless protocol?
For our purposes, a connectionless protocol is asynchronous, and without a persistent connection between client and server. One example of a connectionless protocol would be the world wide web, or Hypertext Transfer Protocol (http). In HTTP communications, a client sends a very small piece of data ( a 'request') to a server, and breaks off the connection. The server responds by sending out the requested file, then ending its connection. In other words, while you are looking at a web page you are not connected to another computer or using any bandwidth until you click on a link, thereby making another request. Connectionless protocols are bandwidth savers, because only while useful information is being exchanged do the client and server occupy bandwidth.
Back to top

What is meant by client/server?
Client/Server software is designed to run on two or more machines at the same time. One machine, which is running the server software, answers requests, often from numerous other 'client' machines with 'client' software. A familiar example to many on the internet is the WWW, in which 'servers' like Apache, NCSA, MS II, Netscape, answer requests from numerous 'clients' like Netscape and MS Internet Explorer. The server and clients do not have to be running on identical machines. For instance, our UNIX web server can interact with clients on MAC, Wintel, UNIX, or any other platforms that can run a WWW client.
Back to top

What is a server?
There are basically two meanings for this word in the internet vernacular.

Hardware: A 'server' can be the word used for the actual piece of hardware (the 'computer') that is running, delivering services to users. (usage: "Our server has a 4.5 gigabyte hard drive and 64 Megs of RAM")
Software: A server is the word used for the piece of software that runs on a particular port, accepting requests from and interacting with remote clients. (usage: "The real audio server has been a nice addition to cyberstation. Do you have the real audio client yet?")

What is a client?
In the language of the internet, a client is a piece of software that is designed to interact with a remote server. (usage: "Netscape is my favorite http client.")
Back to top

What is a shell account?
A so-called 'shell' account is a terminal-type account on our UNIX servers at the office. A shell is basically a program that is interactive and accepts commands from the user. By using telnet, rlogin, or by dialing our modems with your plain user name, you can gain access to your shell account at Cyberstation. When you are in a shell session, your computer acts as a remote terminal on our UNIX machine. The UNIX machine writes on your terminal screen and accepts input from your keyboard. The UNIX shell environment is an extremely powerful internet tool. It is highly programmable and customizable. With your shell account you have internet access at full T1 speeds (the server is directly on the internet). You have access to extensive programming tools, and cron, the system clock daemon.
Back to top

What is a PPP account?
A PPP account is a TCP/IP networking type of account. When you call our modems with a capital P in front of your username, our Terminal Server begins a PPP (point to point protocol) session with your computer. This means that your computer gets a unique IP number, and that traffic destined for your IP number is successfully routed to your machine. When you connect to Cyberstation with a PPP account, it means that your computer is now another 'host' or 'peer' on the internet - able to interact with any other internet computer which gives you permission. Unlike a BBS, shell, or terminal type connection, the PPP connection is NETWORK connection - your computer is not talking directly to another computer, it is just given a network address and routing is begun.
Back to top

What is telnet?
Telnet is a protocol for remote computing on the internet. It allows a computer to act as a remote terminal on another machine, anywhere on the internet. This means that when you telnet to a particular host and port, the remote computer (which must have a telnet server) accepts input directly from your computer (which must have a telnet client) and output for your session is directed to your screen. At cyberstation the primary means of accessing your UNIX shell account is telnet. There are many library and information resources that are accessible through telnet. CyberStation's ftp site has several telnet clients available for download.
How do I use ewan to telnet?
Here is one way for using telnet with ewan, a popular telnet client: The following assumes you already have trumpet Winsock or some other dialer going and are connected to your PPP account.

Open Ewan - it comes up with a connect to site window
choose new (i.e. click on new) - a dialog box pops up
enter cyberstation for the name
enter cyberstation.net for the network address or host name
click ok
you are back on the connect to site screen - click on cyberstation then ok
the terminal screen should come up, and at the bottom it should say connection established
the login prompt for cyberstation will show up in the terminal screen at this point enter your username and password - you are logging onto your shell account on the Unix server at the cyberstation offices.
you will get the Unix command line prompt
it is here that you use the mkdir cd and ls commands as I told you above
to end your shell session, type exit and press enter

How do I use Windows 9x to telnet?
Win9x comes with a built in telnet client. Here is how to use it:

log on to cyberstation using dial-up networking
go to the MS-DOS prompt in Win9x
at the MS-DOS prompt type: telnet cyberstation.net <---(or any other site with a telnet server)
the windows telnet client opens a new window and you get the login prompt
log on to the shell

What is FTP?
FTP stands for File Transfer Protocol. FTP is the best means for moving large files across the internet. FTP is a client/server protocol that enables a user with an FTP client to log on to a remote machine, navigate the file system of that remote machine, and upload and download files from that machine. There are two basic types of FTP on the internet, anonymous ftp and private ftp. With anonymous ftp, one logs in as user anonymous, giving one's email address as a password. With private FTP, one logs in with the username and password one has established on that particular system. You are logged into your home directory, with all the file permissions you would normally have there.
Back to top

What is IRC?
Internet Relay Chat, or IRC, is a protocol for real-time chatting and file transfer on the internet. The largest IRC network is efnet, whi ch usually has as many as 4,000 users on over 1,000 channels active at any one time. You can use IRC from an IRC client like MIRC, or you can use IRC from your UNIX shell account.
Back to top

What is Real Audio?
Real Audio is a client/server protocol designed for streaming audio information in real time over the internet. This provides an internet service like Cyberstation with the ability to broadcast audio over the internet, much like a radio station does over the airwaves. Cyberstation has two Real Audio servers and is expanding its offering of Audio files daily. We enable our business users to serve audio from their web sites as well. To obtain the latest Real Audio client, go to www.realaudio.com.
Back to top

What is TCP/IP?
TCP/IP stands for Transmission Control Protocol/Internet Protocol. It is the standard, basic protocol for all services on the internet. Invented by the Department of Defense, it was first actualized and put in use with the Berkeley UNIX operating system, which was designed under contract with DOD to interoperate using TCP/IP as a basic, native protocol for the BSD UNIX operating system. TCP/IP standards are maintained by the Internet Engineering Task Force (IETF) and described in documents known as RFC's (requests for comments).
Back to top

What is a port?
A TCP or UDP port is what is used by an internet server to distinguish between requests for different services. For instance, traditionally telnet is served at port 23, while a web server listens for requests on TCP port 80. Ports are sometimes shown at the end of a URL; for instance the url telnet://server.com:9900 refers to a telnet server that is running on port 9900 of the host server.com.
Back to top

What is a host? A hostname?
A host is merely the word we use for a computer that is connected to the internet and has its own processor and IP number. A hostname is the name of that computer, and a fully qualified hostname describes the hostname, and the network name, of the computer in question, thereby giving the full path to that host.
For instance: 205.167.0.1 - the IP number of a host
quartz - a hostname for 205.167.0.1
quartz.cyberstation.net - a fully qualified hostname. (i.e. different from, say, quartz.lucky.net)
Back to top

What is an IP address?
An IP number is a unique identifying number that your computer uses during its time on the internet. Traffic coming from you or going to you is routed based on your IP number, which identifies the unique machine you are on and its place in the internet. For instance, 205.167.0.1 is the IP number of host quartz.cyberstation.net. All packets on the internet destined for quartz.cyberstation.net are sent to this IP number by the Domain Name Service. All traffic coming from this host is identified as such because the packets contain the originating IP number in the packet headers. All routing and traffic on the internet ultimately depends on IP numbers.
Back to top

What is a static IP address?
Normally a user is given an unused IP number that is his to use during the duration of that session, and then is put back in the pool of unused numbers when the session is ended. This means that the user will not generally get the same IP number twice. For some users, their application (many times games) requires that they know their IP number, and so they request a 'static' IP number. This means they have a permanent IP number and hostname that they use every time they get on the internet.
Back to top

What is a domain? A domain name?
A domain is a name for a network that is connected to the internet, that has the IP numbers of its hosts mapped to their hostnames by the domain's domain name server. A domain name is the suffix placed on all hosts in that domain. For instance, our domain is cyberstation.net. All of our email addresses show this. All of our hostnames (like quartz.cyberstation.net, dialup9.cyberstation.net) end in the cyberstation.net suffix. We are known by name to the other networks connected to the internet as the domain cyberstation.net. All the computers directly connected to our network are in the domain cyberstation.net.
Back to top

What is a domain name server?
A domain name server, also called a dns, or nameserver, is a machine that handles name queries for all hosts within a particular domain. The domain name server is the authoritative source for addressing information about hostnames within its domain. What this means is that when another computer needs to know the IP number (address) of a host within that domain, its sends a message to the domain name server for that domain, asking that the hostname be resolved and the host's IP number be returned to the machine sending the query. This occurs because all routing and addressing on the internet happens via IP numbers, and not names. Names (like aol.com, cyberstation.net, pathfinder.com) are used for human convenience and all actual addressing is done based on IP numbers. The domain name system is the name for the worldwide distributed database that maps all the world's hostnames to their proper IP number. A domain name server serves hostnames and their IP numbers for a particular domain. Cyberstation has two domain name servers (for backup) one at 205.167.0.5 and one at 205.167.0.1.
Back to top

Who is interNIC?
The interNIC, or Network Information Center, is the central registry for domain name registration in North America. The interNIC was originally funded by the National Science Foundation, but now it is a for-profit entity. The internic takes money for maintaining the root servers, which are servers that answer queries about one of the seven 'root' domains (.com,.net.,edu,.gov,.org,reverse,.mil)
Back to top

What is a gateway?
A gateway is a computer that lies at the intersection of two networks, and routes traffic correctly from one traffic to another, while keeping traffic internal to the two networks seperated. Our gateway, a Cisco 3640 router, lies at the border between cyberstation.net, Qwest, and Cable & Wireless. It correctly routes your traffic bound for the internet out to its destination, while keeping traffic meant for within cyberstation on the cyberstation.net side of the router. Same for Qwest and C&W.
Back to top

What is a netmask?
A netmask is a series of bits designed to 'mask' certain portions of an IP address. The standard netmask for a class C network like cyberstation is 255.255.255.0 - basically the last octet, .0, says 'look here for the machine number, the rest (255) is network number'. The primary use for netmasks is in subnetting.
Back to top

What is MTU and its proper settings?
MTU stands for 'Maximum Transmission Unit' and, as its name implies, is the maximum size of each packet sent (in octets). If you notice that your transfers slow down after time, then adjusting the MTU should help to remedy the problem. An optimal setting for MTU would be 1024 over a modem connection, or 1500 if you are connected to a TCP/IP ethernet network. The normal setting for a Winsock connection will be 576. In any case if you're using Trumpet, the TCPRWIN should be set at 4 times the MTU, and the TCPMSS should be about 40 less than the MTU.
ie. MTU = 296, TCPRWIN = 2048, TCPMSS = 256
or MTU = 1024, TCPRWIN = 4096, TCPMSS = 984
or MTU = 1500, TCPRWIN = 6000, TCPMSS = 1460
Simply go into 'File' and then 'Setup' and make the appropriate changes. If you're using Windows 95 or a Macintosh, you'll need to seek documentation (check out the WWW for this) on how to make the changes.
Back to top

What is an applet?
An applet is a small executable module, that normally doesn't have the complete features and user interface of a normal application. Java is the language most commonly associated with applets. An applet is like a small piece of executable code that needs a full application to contain it. The applet runs inside of the application in a "sand box" or "virtual machine," which is a set of computer resources and instructions that make up an environment for the applet's execution.
Back to top

What is a cookie?
A cookie is a little piece of information handed to a WWW client by a WWW server, that contains information that can be retrieved by the server later. For instance, a server can hand your browser a cookie when you fill out a form stating your preferences for visiting that server. Then, the next time you visit the site, the server can read your preferences from the cookie it put in your browser on the first visit, and customize the appearance of the web site to your preferences. A cookie can also be called a Persistent Client Side State Object.
Back to top

What are frames?
Frames are an extension to HTML invented by Netscape that add much greater functionality, control, and navigability to HTML programming. With frames a WWW programmer can break the main window of the browser into separate panes or frames, each of which can be controlled separately by the programmer. With frames, often you have one frame acting as a 'control frame' while the other is the 'target frame' - the results of actions in the control frame display in the target frame.
Back to top

How do I go back in frames?
The following is paraphrased from Netscape, and merely put here because Netscape is often too busy to respond to users:

HOW THE BACK BUTTON WORKS WITH FRAMES As you may know, the Back button lets you review pages that you've just visited by recalling the URL of an entire screen or frameset. It doesn't, however, call up the URL of the last frame you viewed. So, if you press the Back button when you simply want to revisit a previously chosen frame, you'll find yourself further back than you meant to be. With Netscape 3.0 and later this problem has been fixed.
GOING BACK INSIDE A FRAME (OR, HOW TO GET BACK TO WHERE YOU ONCE BELONGED) Windows and Unix users can click the right mouse button to make the pop-up menu appear. Choose "back" from the menu that appears when you right click inside the frame you want to go back in. Repeating the process will step you back through everything you viewed in that frame in reverse order. If you're using a Macintosh, you can go back one frame by positioning your cursor inside the frame that originally displayed the information you're seeking and holding down the mouse button. A pop-up menu will appear. Choose Back in Frame, and the frame will fill with the information that last appeared in that frame. It's also possible that you'll start out in a frame and follow a link to a window designed without frames. In this case, the frames you've been seeing will disappear. Now, pressing the Back button will lead you back to the last framed page.
BOOK MARKING A FRAME To create a bookmark in any frameset, position your cursor over the link you'd like to bookmark and hold down the mouse button. When the pop-up menu appears, choose Add Bookmark for this Link. Please note that if you choose to create a bookmark for the entire page, it will bookmark the URL for the original frameset, which may not be the bookmark you meant to choose.
PRINTING A PAGE To print a page, select the frame you want to print by clicking in it. Then, from the File menu, select Print Frame.

What is a home directory?
Your "home" directory is the directory (or "folder") on CyberStation's UNIX servers that contains all your personal files, some configurations, your web space, etc. It's the place you are immediately after you log into the system with telnet. The name of this directory is the same as your login and the user in his or her own home directory has complete control over all files and permissions on files found there. Because our UNIX system is a multi-user system, unlike a home computer where one person is presumed to "own" everything, the files and directories on the system are all associated with a particular user, or "owner." You own your home directory and all the files and directories under that. This gives you your own private directory tree that is under your control on the server.
Back to top

What is a pubhtml directory?
Your pubhtml directory is made available on the web as url http://www.cst.net/~username/. It is a directory automatically created as a subdirectory of your home directory. Your web pages, cgi-bin directory, and other files and directories for the web go there.
Back to top

What are UNIX permissions and how do I set them?
Permissions are what give a user control over his or her files. The user can control who has access to his files, by setting the permissions accordingly. Permissions of some kind are really necessary on a multi-user system. You wouldn't want just anyone looking at your mail, or other sensitive files. This is controlled with permissions. On the other hand, if permissions on files that are supposed to be public (like web pages) are 'cut off' or have the 'world readable' permission set to 'no' then folks can't see your web page. There are two ways to set permissions in UNIX, symbolically, and numerically. Either method works at cyberstation. You also need to know how to display access permissions.
Back to top

How can I use Cyberstation's Windows 9x/UNIX integration to access my server directory with Windows networking?
CyberStation uses the SAMBA protocol to give our users access to the UNIX internet server directories just as though they were on a Windows95 network. Here is how to use this feature of Cyberstation:

Your Networking and Dial-Up Connection to Cyberstation must be setup as outlined in Win95 Settings.
Establish a connection via dbl-clicking the Cyberstation Dial-Up connection.
Now, you can map your network drive one of 2 ways:

A. Desktop - Right click on the 'My Computer' icon on the desktop and select 'Map Network Drive'.
B. Windows Explorer

Open Windows Explorer and make sure that you see the toolbar (if not: click the view pull-down menu and check Toolbar).
Select the 3rd button which is 'Map Network Drive'.

No matter which method you use, then dialog window will be as shown below. Win95 automatically selects the next available drive letter, or you may select one. Enter the path with '\\quartz\<your username here>.

As to whether or not you want to map the network drive every time you logon to your computer or even for Win95 to save your password is up to you. After completing this, the screen should refresh and you will have access to your drive space on the web server.

CST.net Technical Questions

More Advanced Questions

ADs