comparison thesis/tex/3-MailTransferAgents.tex @ 126:27ddf2506157

outsourced floats; minor stuff
author meillo@marmaro.de
date Tue, 09 Dec 2008 16:04:02 +0100
parents 1cb6a2f5f077
children 6f622eb5c812
comparison
equal deleted inserted replaced
125:aa1fb227e68e 126:27ddf2506157
66 66
67 67
68 \section{Popular MTAs} 68 \section{Popular MTAs}
69 69
70 %todo: include market share analyses here 70 %todo: include market share analyses here
71 << some info about market shares >>
71 72
72 One would not use a program for a job it is not suited for. Therefor only \mta{}s that are mostly similar to \masqmail\ are regarded here. These are \emph{sendmail-compatible} ``smart'' \freesw\ \MTA{}s that focus on mail transfer. 73 One would not use a program for a job it is not suited for. Therefor only \mta{}s that are mostly similar to \masqmail\ are regarded here. These are \emph{sendmail-compatible} ``smart'' \freesw\ \MTA{}s that focus on mail transfer.
73 74
74 For the comparison, five programs are taken: \sendmail, \name{exim}, \name{qmail}, \name{postfix}, and \masqmail. The four alternatives to \masqmail\ are the most important representatives of the regarded group. % FIXME: add ref that affirm that 75 For the comparison, five programs are taken: \sendmail, \name{exim}, \name{qmail}, \name{postfix}, and \masqmail. The four alternatives to \masqmail\ are the most important representatives of the regarded group. % FIXME: add ref that affirm that
75 76
76 \name{courier-mta} is also a member of this group, being even closer to \name{groupware} than \name{postfix}. It is excluded here, because the \NAME{IMAP} and webmail parts of the mail server suite are more in focus than its \MTA. Common mail server setups even bundle \name{courier-imap} with \name{postfix}. %fixme: need this sentence? 77 Other members are: \name{smail}, \name{zmailer}, \name{mmdf}, and \name{courier-mta}; they all are less important and rarely used.
77
78 Other members are: \name{smail}, \name{zmailer}, \name{mmdf}, and more; they all are less important and rarely used.
79 78
80 Following is a small introduction to each of the five programs chosen for comparison, except \masqmail\ which already was introduced in chapter \ref{chap:introduction}. 79 Following is a small introduction to each of the five programs chosen for comparison, except \masqmail\ which already was introduced in chapter \ref{chap:introduction}.
81 80
82 81
83 82
138 137
139 138
140 139
141 \section{Comparison of MTAs} 140 \section{Comparison of MTAs}
142 141
143 << general fact in table \ref{tab:mta-comparison} >> 142 This section tries not to provide an overall \MTA\ comparison, because this is already done by others: Including
144 143
145 Refer to \cite{hafiz05}. 144 \url{http://shearer.org/MTA_Comparison}
145 \url{http://www.geocities.com/mailsoftware42/}
146 \url{http://fanf.livejournal.com/50917.html}
147 \url{http://archives.neohapsis.com/archives/postfix/2006-07/1762.html}
148
149
150 For a discussion on \mta\ architectures (comparing \sendmail, \name{qmail}, \name{postfix}, and \name{sendmail X}) it is refered to Hafiz \cite{hafiz05}.
151
152 Here provided is an overview on a selection of important properties, covering the four previously introduced programs. Table \ref{tab:mta-comparison} provides it.
153
146 154
147 \begin{table} 155 \begin{table}
148 \begin{tabular}[hbt]{| p{0.13\textwidth} || p{0.13\textwidth} | p{0.13\textwidth} | p{0.13\textwidth} | p{0.13\textwidth} | p{0.13\textwidth} |} 156 \begin{center}
149 \hline 157 \input{input/mta-comparison.tex}
150 158 \end{center}
151 & sendmail & exim & qmail & postfix & masqmail \\ 159 \caption{Comparison of MTAs}
152 \hline \hline 160 \label{tab:mta-comparison}
153 First release & 1983 & 1995 & 1996 & 1999 & 1999 \\
154 \hline
155 Lines of code (with sloccount on debian packages)& 93k & 54k & 18k & 92k & 14k \\
156 \hline
157 Architecture & monolithic & monolithic & modular & modular & monolithic \\
158 \hline
159 Design goals & flexibility & general, flexible \& extensive facilities for checking & security & performance and security & for non-permanent Internet connection \\
160 \hline
161 Market share (by Bernstein in 2001) & 42\% & 1.6\% & 17\% & 1.6\% & (unknown) \\
162 \hline
163
164 \end{tabular}
165 \caption{Comparison of MTAs}
166 \label{tab:mta-comparison}
167 \end{table} 161 \end{table}
168 162
169 163
170 \subsection{about market share} 164 \subsection{about market share}
165 \url{http://www.oreillynet.com/lpt/a/6849}
166
167 \url{http://www.mailradar.com/mailstat/}
168
169 Market share (by Bernstein in 2001): sendmail 42\% , exim 1.6\% , qmail 17\% , postfix 1.6\%.
170 masqmail has no relevant market share (debian popcon)
171
172
173
171 174
172 \subsection{About architecture} 175 \subsection{About architecture}
173 176
174 \subsection{Security comparison} 177 \subsection{Security comparison}
175 178
176 179
177 \url{http://shearer.org/MTA_Comparison}
178
179 \url{http://www.geocities.com/mailsoftware42/}
180
181 \url{http://fanf.livejournal.com/50917.html}
182
183 \url{http://archives.neohapsis.com/archives/postfix/2006-07/1762.html}
184
185 \url{http://www.oreillynet.com/lpt/a/6849}
186
187 \url{http://www.mailradar.com/mailstat/}
188 180
189 181
190 182
191 183
192 184
207 199
208 << community (amount and quality) >> 200 << community (amount and quality) >>
209 201
210 << used it myself >> 202 << used it myself >>
211 203
212 << had problems with it >> 204 << had problems with it >>
213 205
214 206
215 207
216 208
217 << quality criteria >> %FIXME 209 << quality criteria >> %FIXME
224 216
225 << how many criteria for ``good''? >> %FIXME 217 << how many criteria for ``good''? >> %FIXME
226 218
227 219
228 220
229 << from the practice of programming: are the names good? check the significant number of characters. (intern: 31char, extern: 6char caseless; ProgC p.184) >> 221
222
223 Ref back to \ref{sec:what-will-be-important}:
224
225 provider indepencence -> easy config:
226 \sendmail\ and \name{qmail} appear to have bad positions at this point. Their configuration is complex, thus they would need simplification wrappers around them to provide easy configuration.
227
228 performance not so important:
229 \name{postfix} focuses much on performance, this might not be an important point then.
230
231 security:
232 It seems as if all widely used \mta{}s provide good security nowadays. \name{qmail}'s architecture, also used in \name{postfix}, is generally seen to be conceptually more secure, however.
230 233
231 234
232 --- 235 ---
233 236
234 But for example delivery of mail to local users is \emph{not} what \mta{}s should care about, although most \MTA\ are able to deliver mail, and many do. (\name{mail delivery agents}, like \name{procmail} and \name{maildrop}, are the right programs for this job.) 237 But for example delivery of mail to local users is \emph{not} what \mta{}s should care about, although most \MTA\ are able to deliver mail, and many do. (\name{mail delivery agents}, like \name{procmail} and \name{maildrop}, are the right programs for this job.)