Tutorial Part Two: Frequently Asked Questions
1. What is the relationship between
an email address and a URL?
Universal Resource Locator (a.k.a. URL) is
the term for an “address” for a specific location on the Internet, much
like a ZIP code points to a location in the “real world” mail system.
Below is an example of a URL, as you might use
your web browser:
The URL is broken down into three basic components:
Protocol
The Protocol directs the browser on how to connect to the machine specified in
the rest of the URL. The above example calls for HTTP (HyperText Transfer
Protocol) HTTP is the standard language used for sending web pages. Other
protocols can be used, such as FTP (File Transfer Protocol) used in the transmission of
files.
Machine Name This component specifies the
name of the machine where the web server is locate
within the tekguard.com domain.
Domain Name - This component specifies
the Domain Name. It is the key most element in determining
the destination (location of the machine to be browsed.)
2. How is email transferred between
the Client and the Mail Server?
When the Email Client sends a message, it first finds the
Mail Server using the Domain Name Systems (DNS)*. Contact with the Mail
Server is accomplished using the “Simple Mail Transfer Protocol (SMTP).
After the Client contacts the Mail Server,
the message is then sent; the Email Client will then send any other pending
messages or if it is done it will log off from the Mail Server.
The collection of email messages by the
Email Client is similar. The Client uses the “Post Office Protocol”
(POP) to send an account name (user name) and a password to the Mail Server so
it knows for whom to collect outstanding messages, which are then collected
one at a time.
Note: On a local network there may not be
a DNS server; in which case the Mail Server is referred to by its actual IP
address rather than by a Domain Name.
3. What is Simple Mail Transfer
Protocol (SMTP)?
Simple Mail Transfer Protocol (SMTP) was
developed during the 1980 as a standard set of rules and procedures whereby
messages could be transferred over the Internet. It is the common
language used by the majority of Mail Servers to send messages back and forth
to other Mail Servers or Email Clients.
SMTP is a set of simple commands that are
sent to the Mail Server. These commands include the sender’s email
address, the recipient’s email address and the actual data being sent (your
message). SMTP is so ubiquitous that your Email Client will
automatically use this protocol without your direct knowledge. Below is
a list of sample SMTP commands:
HELO post.tekguard.com Says “hello” and gives the name of the Mail Server sending the message.
RCPT TO <irene@company1.com> States to whom the message is being sent.
MAIL FROM <sales@tekguard.com> States from whom the message was sent
The SMTP system uses Domain Name System
(DNS) to convert the Domain Name component of a typical email address, to the
specific network address of the machine that administers the Domain. It
then establishes a communications link with Mail Server and transfers the
email message.
Since the DNS system specifies the
recipient’s actual machine address, SMTP is useful only where the receiving
machine is always at the same address.
4. What is Post Office Protocol (POP)?
In order to collect email messages from the
Mail Server, the Email Client contacts the Mail Server and supplies it with an
account “user” name and a password. Only then will the Mail Server
offer a count of pending messages stored for that account. Armed with
the quantity of messages awaiting it, the Email Client requests each message,
one at a time until all of the messages have been retrieved. After the
Client application has downloaded the messages, they will be either deleted
from the Mail Server or marked as having been read, according to how the Mail
Server has been configured.
Mail Servers are usually configured to limit
the maximum quantity of stored messages for any single account; whether the
messages have already been collected or are still waiting to be read.
The Email Client collects messages from the
Mail Server according to the rules and procedures known as the Post Office
Protocols (POP). The current version of POP is POP3. Mail systems adhering
to this standard are referred to as POP3 mailboxes. Some Email Clients refer
to POP3 simply as POP. A POP3 mailbox can also be referred to as a POP account.
Older protocols such as POP2 are still in existence but are quickly falling
out of favor and use.
POP3 communication can be initiated from
any address and can be used to collect mail over part time links or any connection
to the Internet.
To collect mail from an account, POP3 first
expects an account name and password to be supplied. The commands are
similar to what we’ve already seen in SMTP. POP3 also tends to be used
automatically so you might never notice what the commands look like.
Below are a few examples:
USER Eric States
your account name.
PASS Eric States your password.
STAT Reports the number of messages awaiting retrieval in that account.
RETR 1
Retrieve message #1 currently stored in the account.
DELE 1 Delete message #1 currently stored in the account.
5. What is the relationship
between POP3 and SMTP?
Except for the special cases listed below;
POP3 is used to collect email messages from a Mail Server and STMP is used for
more extensively in the sending of email messages.
* SMTP is occasionally used in the
collection of email messages from systems that do not support POP3. In
such rare cases the Email Client is made to mimic the functionality of a Mail
Server. This is accomplished via an intermediate application accessed by
the Email Client for such a purpose.
** POP3 is occasionally called upon to collect mail
instead of STMP from a Mail Server or by some other server in certain special
cases; such as when one of the Mail Servers is on a part time connection to
the Internet. See the FAQ section below What do I need to do to receive
mail via a Mail Server if I have a Dynamic Address?
6. What is DNS or a DNS Server??
The Internet works on IP addresses; DNS (Domain Name
Services) Servers are used to convert the addresses we see and read into IP
addresses and vice-versa. It would be very hard to remember an address
such as: 127.0.0.7, which is the IP address equivalent of our old friend; “www.tekguard.com”.
In addition to the immediate and obvious
benefit of using an easily recognizable and memorable Domain Name, there is a
more substantial benefit. Let’s say that the address changes for a
network server or mail server, you would only have to change the DNS record
for that specific machine on the DNS server. Then all the other machines
that need to access the mail or network server would be able to do so without
the need to inform them of the change of IP address! In this instance
the DNS server functions like a “forwarding service” or “clearing house”
making the change of address transparent to all the other machines.
The specific information for every machine
within a particular Domain (i.e. “tekguard.com”) is stored in a simple
database; each machine having its own entry in the database called a DNS
Record. Below is an example or a DNS Record:
The DNS Record includes the specific name
of the machine within a particular Domain as well as its IP address. The
above example is the DNS Record for TekGuard’s web server. The IP
address is what a web browser actually uses to contact the Web Server, rather
than the name “tekguard.com”.
If you wanted you could simply input the IP
address for www.tekguard.com into your browser and the TekGuard web site would
also be brought up!
There can be much more information stored
in a DNS record, but as the scope of this FAQ is more limited; please refer to
the list of URL’s at the end of this section for further reading on DNS.
7. How do DNS and DNS Servers relate
to email?
SMTP requires the use of the DNS to convert
the Domain Name portion of an Email Address into the real location of the
machine that is actually running the Mail Server that administers the Email
functions for that particular Domain.
Special entries in the DNS Server called MX
(Mail eXchanger) Records are used to accommodate Email communication.
A typical MX Record is shown below:
Domain
Name
The Domain Name is the name that an entity has registered for their exclusive
use on the Internet. In the above example VInfo had registered TekGuard.com as the Domain Name for their own use on the Internet.
Priority
When there are multiple MX Records, Email Server and Email Clients will use
the Priority designation to determine which record is accessed. The
lower the Priority number the higher the actual priority placed on that
specific Record. The Priority is a relative value. The actual
number is of no real significance since priority is established by the
relative rank of the MX Record compared to other MX Records.
Mail
Server
This portion of the MX Record states the name of the machine which is running
the Mail Server application that administers the email on a Domain.
In this example “post.tekguard.com” is
the machine that administers the email for the “tekguard.com” domain.
8. How can people send me email when my Mail Server is
not on a permanent link to the Internet?
The Mail Server maintained by your ISP will store email addressed to your own
Domain until your Mail Server can connect to the Internet. Two separate
MX Records are kept on the DNS Server to facilitate just such a
situation. In the below example the following MX Records are assumed:
In this example, Irene is sending email to Mark.
Irene’s Email Client first sends the message to the Mail Server for “company1.com”.
The Mail Server for “company1.com” will
then research where it will need to send the message in order for Mark to get
it. This is done by looking up an MX Record on the DNS Server. In
simple terms, the Mail Server at “company1.com” looks for the machine that
administers email for the Domain “company2.com”.
The DNS Server contains two MX Records for
the Domain “company2.com”. Irene’s Mail Server will determine
which machine to attempt delivery to first by comparing the Priority value on
each MX Record. The machine specified in the MX Record with the smaller
Priority value will be first to receive the message. For our example the
MX Record lists “post.company2.com” as having the lower Priority Value.
If the Mail Server for “company2.com”
is currently connected to the Internet, then the message is transferred
directly to it without ever going through an intermediate Mail Server.
However, let’s say the “company2.com” server is off-line. The
first attempt to deliver the message will fail and the “company1.com” mail
server will use the next MX Record available for “company2.com”, which is
“relay.myisp.com”. Since this Mail Server has a permanent live
connection to the Internet the mail will be transferred there.
The “relay.myisp.com” Mail Server will
try to forward the message to the Mail Server with the lowest MX Record
Priority value on a regular interval. If the “company2.com” Mail Server is off-line for
some time, “relay.myisp.com” will send the message to the machine with the
next higher Priority value listed on the MX Record. Since the next
machine is itself, it will suspend delivery for a time before attempting
another delivery. This is repeated until the message can be forwarded to
“post.company2.com”, but after a certain number of attempts or a certain
amount of time the “relay.myisp.com” Mail Server may give up,
consider the
message undeliverable and return the message to the original sender with an
error message.
However, if the “post.company2.com”
Mail Server reestablishes an Internet connection the message will be
transferred and Mark’s Email Client will then be able to retrieve the
message for him to read.
9. How do I transfer email from my
ISP's Mail Server to my own Mail Server?
There should be no concern about triggering your ISP Mail
Server to send mail if you are transferring the email messages to your own
Mail Server using POP3 protocols; because your Mail Server will collect it
at a predetermined time automatically. Although this is an easy means of
collecting messages there can be hidden problems. They are discussed in more
detail in Section 10.
If you are not using POP3, then your Mail Server will
passively wait for messages to come in. Your own Mail Server will rely
on other Mail Servers to send it messages at their own leisure as opposed to
triggering them to send any outstanding messages. Although ISP’s offer
different services, the following methods of mail recovery will address your
most common options:
1.
At specific intervals our ISP’s Mail Server will determine if your
own Mail Server is available. This method is commonly used for Mail
Servers permanently connected to the Internet, utilizing MX Records as
discussed in Sections 7 & 8.
The disadvantage to this passive method of mail collection is that your ISP’s
Mail Server may not check to see if your own Mail Server is on-line during
those times that it is actually on-line! If your ISP sets its Mail
Server to check previously unavailable Mail Servers only sporadically they you
may not get your email messages in reliable, timely manner. The typical
“re-try” interval is “every 30 minutes” and that may too long of an
interval for users who are on temporary, dial-up connections.
2.
During the times that your own Mail Server is connected to the
Internet, it send an email message to the ISP’s Mail Server declaring it is
available to accept stored email. This will trigger a flow of messages
to your own Mail Server. This method is known as “wake up mail” and
it was instituted to help in situations where the Mail Server is not always
“on-line”. However it is not widely used since the next method
supersedes it.
3.
Using the “remote queue starting” or ETRN commands, your own Mail
Server “tells” the ISP’s Mail Server that it is available to accept
messages, which then triggers the flow of stored messages to your own Mail
Server. This method is not supported by all Mail Server but since it was
recently made an Standard protocol it should become more widespread.
4.
Your own Mail Server “fingers” the ISP’s Mail Server in a process
similar to the ETRN method mentioned above. A “finger” is a type of
probe that is sent and request information.
With all of the above methods, while you are connected to the Internet, you
will receive mail directly from senders and you will also be able to send mail
directly to your ultimate recipients. However you may choose to use your
ISP’s Mail Server as an intermediary for both incoming and outgoing mail
deliveries. The benefit being that you can be online for only short
periods of time and still get mail in and out without risking any delays if
you were trying to deliver mail to unresponsive or busy Mail Servers.
With this arrangement your ISP’s Mail Server (which is on a permanent
connection) can worry about making contact with troublesome destination Mail
Servers.
10. What are common POP3 collection
problems?
Although the POP3 method of email
collection from an ISP’s Mail Server is generally simple, there are a few
common problems that can occur.
Messages are stored at the ISP’s Mail
Server in a single account for your entire Domain. Occasionally the
address for the final recipient of the message may be lost. This happens
especially when the recipient was addressed as a “BCC” or Blind Carbon (or
Courtesy) Copy, since the address for this type of recipient is not listed as
part of the header.
As this strategy of addressing is becoming
more common with the increase of electronic mailing lists, the number of “undelivered
messages” is also on the rise. These messages end up on your Mail
Server and are undeliverable. The administrator for your Mail Server
will have to manually examine these messages to determine their ultimate
recipient. If your organization is a large one this task could end up
being a daunting one!
If you find that the number of messages lost
because they are sent BCC is unacceptably high and the Mail Server Administrator
cannot cope with manual delivery, then you may want to consider the following
two measures:
1.
Your own Mail Server may offer you the option of using “custom mail
headers” which specify recipient information. Your ISP will be
required to add a special line at the top of each message headed to your
Domain, specifically stating the recipient’s name. Your mail server
will use this top line to deliver all messages including BCC’s correctly.
Of course, you will need to determine if your Mail Server gives you this
option and if your ISP is willing to add a special section to each of your
inbound messages
2.
Another solution is to inform
your Mail Server administrator which “mail lists” you belong to so the
Mail Server can be set up with a special rule to direct mail coming from those
specific sources directly to you without relying on your own address to be
read from the incoming mail.
This method will not require your ISP to do anything special, but
unfortunately mail from unknown (or not listed) senders will still end up in
the “lost messages” file and will need to be processed manually by your
Mail Server administrator, although the quantity will be greatly reduced.
11. What else should I be concerned
about in setting up an email system?
If you chose to set up a Mail Server in your
own office with the various Email Clients processing their incoming and
outgoing mail through it, then you must have that Mail Server exist on two
different networks. The one network will be for communication with the
various Email Clients and the other network will be for communication via the
internet to other Mail Servers, such as your ISP’s Mail Server.
The following guidelines will help you set
up your Mail Server so it exists on separate networks:
A.
Your Mail Server must be configured correctly to use TCP/IP and to send
and receive email messages via the Internet. The various Email Clients
will then communicate with the Mail Server using another protocol to TCP\IP;
such as Microsoft NetBeui and the Email Clients will use NetBeui to
communicate, but if your system will also accommodate PC’s then they will
have to be modified to use NetBeui.
B.
You can change the TCP/IP configuration on your Mail Server machine
manually, by setting one configuration so you can connect to the Internet to
receive and send messages. Then you will switch to the second
configuration which will allow the Email Client machines to receive and send
messages to the Mail Server machine.
C.
You can operate a router to connect your network to the Internet.
This will allow the Email Client machines to communicate with Mail Server via
the TCP/IP. The Mail Server is then free to communicate to the Internet
through the router.
D.
You can VInfo Email Interceptor (as a Message Transport Agent) on the
same machine you are running your Mail Server on, allowing the Mail Server to
use TCP/IP with both the local Email Clients and on the Internet without
having to change the TCP/IP configurations on your Mail Server machine.
12. I just need one or two email accounts,
what do I need to do?
VInfo’s Email Interceptor Mail Server,
Personal Edition is ideal for a small number of email addresses. Please see
our data sheets for further information.
13. I need to have more than two email
accounts, what do I need?
VInfo's Email Interceptor Mail Server,
Enterprise Edition is designed for small and large businesses requiring a
medium to large number of email addresses. Please see our data sheets for
further information.
14. How do I set up a system where my
email accounts are only at my ISP and not on a local Mail Server?
ISP’s
can allow any number of email accounts free of charge. They may also be
able to supply you with additional quantities if upon request. There may
be an extra charge for setting up and administering your email accounts.
The
advantage of having your ISP host your email accounts is the you don’t need
to know about or worry about administering your own Mail Server.
The main
disadvantage is that each person in your organization will have to connect
to the ISP for email service and that will require some sort of Internet connection
for every machine in your office.
15. What will I need to run a Mail Server
at my office?
You will
need to choose a Mail Server application that is compatible with the type
of Internet account provided by your ISP.
If your
ISP has provided you with a Static Address, then you can download your email
from your ISP with a Mail Server that supports SMTP. If you plan on
operating a system that is not always online, then your ISP will also have to
set up a system to store your email collected during the time your system is
not available for deliveries.
If you
have a Dynamic Address, then you will need a Mail Server application that
supports “Domain Wide Accounts” as well as POP3 collection from your
ISP. You will need to see if your ISP can accommodate this sort of
set-up.
16. What is a “Web Based Mail
Service”?
There are
companies (such as AOL, Yahoo, MSN, etc.) offering you email service via your
Internet web browser (Netscape Navigator or Internet Explorer). To take
advantage of Web Based Mail Service you will have to “log in” to a Mail
Server which will display your mail in your browser window. Web Based
Mail Services offer great flexibility; you do not need to maintain a Mail
Server of your own or use an Email Client; additionally you do not need to use
email addresses that relate to your specific company (no need to maintain a
Domain.)
The burgeoning
growth in companies offering Web Base Mail Service requires that you diligently
compare available services, like forwarding mail to another email account
and hidden costs.
Additional
drawbacks of Web Based Mail Services are:
You
will need to be connected to the Internet to send, receive and even compose
you email messages. As connection time can be expensive you may consider
composing your messages while you are off-line using a text editor and then
copy & paste your message into the web browser when you are actually on-line.
This can be a slow, tedious and inconvenient process.
The
Domain portion of your email address will be that of your Mail Service
provider (i.e. “yahoo.com” if you are using Yahoo for Mail Service.)
Some Web Based Mail Services will allow you to request your own Domain or
choose from a list of available Domains. A charge may required for such
a custom service.
Functionality
may be limited in Web Based Mail Services: such as storing, sorting and searching
old messages. Additional services may also require additional charges.
Sending
and receiving attachments may be limited according to size if available at
all.
May
Web Based Mail Services subsidize their companies by selling advertising.
Your mail box may be stuffed with advertisements and little useful mail.
You
may be limited to a specific browser, whether it be Navigator or Explorer
or your access to mail may be hampered; especially if your browser is set
up not to accept Cookies (a small piece of code sent by web servers to browsers
that supply specific information.)
If
your Web Based Mail Service provider is not local; you may occasionally find
“web congestion” causing problems for you in collecting email.
Security
over the Internet could be lower than for email through your own local Mail
Server. Messages will have to traverse the reaches of the Internet even if
they are just headed to the person in the next cubicle. Collection of mail
would also have to manual performed since forwarding to your web browser is
not an option. If you neglect or forget to check for mail and important message
could go unnoticed for some time.
|
