docs/diploma

annotate thesis/tex/4-MasqmailsFuture.tex @ 196:b80438534651

find changesets by author, revision, files, or words in the commit message
rewrites and new content about non-functional requirements
author meillo@marmaro.de
date Wed, 31 Dec 2008 14:00:30 +0100
parents 74a6cbdc7255
children b08be036783d
rev   line source
meillo@109 1 \chapter{\masqmail's present and future}
meillo@93 2
meillo@185 3 This chapter \dots %fixme write text here
meillo@185 4
meillo@185 5
meillo@185 6
meillo@185 7
meillo@137 8 \section{Existing code base}
meillo@185 9
meillo@142 10 Here regarded is version 0.2.21 of \masqmail. This is the last version released by Oliver \person{Kurth}, and the basis for my thesis.
meillo@142 11
meillo@93 12
meillo@185 13 \subsubsection*{The source code}
meillo@185 14
meillo@185 15 \masqmail\ is written in the C programming language. The program, as of version 0.2.21, consists of 34 source code and eight header files, containing about 9,000 lines of code\footnote{Measured with \name{sloccount} by David A.\ Wheeler.}. Additionally, it includes a \name{base64} implementation (about 300 lines) and \name{md5} code (about 150 lines). For systems that do not provide \name{libident}, this library is distributed as well (circa 600 lines); an available shared library however has higher precedence in linking.
meillo@185 16
meillo@185 17 The only mandatory dependency is \name{glib}---a cross-platform software utility library, originated in the \NAME{GTK+} project. It provides safer replacements for many standard library functions. It also offers handy data containers, easy-to-use implementations of data structures, and much more.
meillo@185 18
meillo@185 19 With \masqmail\ comes the small tool \path{mservdetect}; it helps setting up a configuration that uses the \name{mserver} system to detect the online state. Two other binaries get compiled for testing purposes: \path{readtest} and \path{smtpsend}. All three programms use \masqmail\ source code; they only add a file with a \verb+main()+ function each.
meillo@185 20
meillo@185 21 \masqmail\ does not provide an interface to plug in modules with additional functionality. There exists no add-on or module system. The code is only separated by function to the various source files. Some functional parts can be included or excluded by defining symbols. Adding maildir support at compile time, means giving the option \verb+--enable-maildir+ to the \path{configure} call. This preserves the concerning code to get removed by the preprocessor. Unfortunately the \verb+#ifdef+s are scattered through all the source, leading to source code that is hard to read.
meillo@185 22 %fixme: refer to ifdef-considered-harmful ?
meillo@185 23
meillo@185 24
meillo@185 25
meillo@137 26 \subsubsection*{Features}
meillo@93 27
meillo@142 28 \masqmail\ accepts mail on the command line and via \SMTP. Mail queueing and alias expansion is supported. \masqmail\ is able to deliver mail to local mailboxes (in \name{mbox} or \name{maildir} format) or pass it to a \name{mail delivery agent} (like \name{procmail}). Mail destinated to remote locations is sent using \SMTP\ or can be piped to commands, being gatesways to \NAME{UUCP} or \NAME{FAX} for example.
meillo@93 29
meillo@142 30 Outgoing \SMTP\ connections feature \SMTP-\NAME{AUTH} and \SMTP-after-\NAME{POP} authentication, but incoming connections do not. Using wrappers for outgoing connections is supported. This offers a two way communication through a wrapper application like \name{openssl}.
meillo@137 31 %todo: what about SSL/TLS encryption?
meillo@93 32
meillo@142 33 \masqmail\ focuses on non-permanent online connections, thus a concept of online routes is used. One may configure any amount of routes to send mail. Each route can have criterias, like matching \texttt{From:} or \texttt{To:} headers, to determine if mail is allowed to be sent using it. Mail to destinations outside the local net gets queued until \masqmail\ is informed about the existance of a online connection.
meillo@142 34
meillo@137 35 The \masqmail\ executable can be called under various names for sendmail-compatibility reasons. This is organized by symbolic links with different names pointing to the \masqmail\ executable. The \sendmail\ names are \path{/usr/lib/sendmail} and \path{/usr/sbin/sendmail} because many programs expect the \mta\ to be located there. Further more \sendmail\ supports calling it with a different name instead of supplying command line arguments. The best known of this shortcuts is \path{mailq}, which is equivilent to calling it with the argument \verb+-bq+. \masqmail\ recognizes the names \path{mailq}, \path{smtpd}, \path{mailrm}, \path{runq}, \path{rmail}, and \path{in.smtpd}. The first two are inspired by \sendmail. Not implemented is the name \path{newaliases} because \masqmail\ does not generate binary representations of the alias file.\footnote{A shell script located named \path{newaliases}, that invokes \texttt{masqmail -bi}, can provide the command to satisfy other software needing it.} \path{hoststat} and \path{purgestat} are missing for sendmail-compatibility.
meillo@109 36 %masqmail: mailq, mailrm, runq, rmail, smtpd/in.smtpd
meillo@109 37 %sendmail: hoststat, mailq, newaliases, purgestat, smtpd
meillo@109 38
meillo@137 39 Additional to the \mta\ job, \masqmail\ also offers mail retrieval services with being a \NAME{POP3} client. It can fetch mail from different remote locations, dependent on the active online route.
meillo@109 40
meillo@137 41
meillo@132 42
meillo@132 43
meillo@132 44
meillo@132 45
meillo@177 46 \section{Requirements}
meillo@146 47
meillo@185 48 This section identifies the requirements for a modern \masqmail. Most of them will apply to modern \MTA{}s in general.
meillo@185 49
meillo@185 50
meillo@177 51
meillo@177 52 \subsection{General requirements}
meillo@146 53
meillo@189 54 Here follows a list of quality requirements for \masqmail, or other kinds of programs in similar environment and with similar jobs. These requirements specify the non-functional properties of the software, thus they are also called \name{non-functional requirements}. The list is based on \person{Hafiz} \cite[page~2]{hafiz05}, with insperation from \person{Spinellis} \cite[page~6]{spinellis06}.
meillo@185 55 %fixme: refer to ch01 and ch02
meillo@146 56
meillo@146 57
meillo@177 58 \subsubsection*{Security}
meillo@196 59 \MTA{}s are critical points for computer security, as they are accessable from external networks. They must be secured with high effort. Properties like high priviledge level, work load influenced from extern, work on unsafe data, and demand for reliability, increase the need for security. \masqmail\ needs to be secure enough for its target field of operation.
meillo@177 60
meillo@177 61
meillo@177 62 \subsubsection*{Reliability}
meillo@196 63 Reliability is the second essential quality property for an \MTA. Mail for which the \MTA\ took responsibility must never get lost. The \MTA\ must not be \emph{the cause} of any mail loss, no matter what happens. Unreliable \mta{}s are of no value.
meillo@177 64
meillo@189 65
meillo@189 66 \subsubsection*{Robustness}
meillo@196 67 Being robust means handling errors properly. Small errors may get tolerated, large errors may kill a process. Killed processes should be automatically restarted and lead to a clean state again. Log messages should be written in every case. Robust software does not need a special environment, it creates the right environment itself. \person{Raymond}'s \name{Rule of Robustness} and his \name{Rule of Repair} are good descriptions.\cite[pages~18--21]{raymond03}
meillo@177 68
meillo@177 69
meillo@177 70 \subsubsection*{Extendability}
meillo@196 71 Modern needs like large messages demand for more efficient mail transport through the net. Aswell is a final solution needed to defeat the spam problem. New mail transport protocols seem to be the only good solutions for both problems. They also can improve reliability, authentication, and verification issues. \masqmail's architecture needs to be extendable, to allow new features to be added afterwards. For example new mail transfer protocols as they appear and are used.
meillo@196 72
meillo@146 73
meillo@189 74 \subsubsection*{Maintainability}
meillo@196 75 Maintaining software takes much time and effort. \person{Spinellis} guesses ``40\,\% to 70\,\% of the effort that goes into a software system is expended after the system is written first time.'' \cite[page~1]{spinellis03}. This work is called \emph{maintaining}. Hence making software good to maintain is effort that will easy 40 to 70\,\% of the work afterwards.
meillo@146 76
meillo@189 77
meillo@189 78 \subsubsection*{Testability}
meillo@196 79 Good testability make maintainance easier too, because functionality is directly verifiable when changes are done, thus removing the uncertainty. Modularized software makes testing easier, because parts can be testet without external influences. \person{Spinellis} sees testability as a subquality of maintainability.
meillo@189 80
meillo@189 81
meillo@189 82 \subsubsection*{Performance}
meillo@196 83 Also called ``efficiency''. Efficient software requires few time and few resources. The merge of communication hardware and its move from service providers to homes and to mobile devices, demand smaller and more resource-friendly software. The amount of mail will be lower, even if much more mail will be sent. More important will be the energy consumption and heat emission. These topics increased in relevance during the past years and they are expected to become more central.
meillo@146 84
meillo@146 85
meillo@189 86 \subsubsection*{Availability}
meillo@196 87 Availability is important for server programs. They must stay operational, even during \name{denial of service} attacks.
meillo@146 88
meillo@146 89
meillo@189 90 \subsubsection*{Portability}
meillo@196 91 Source code that compiles and runs on various operationg systems is called portable. Portability can be achieved by using standard features of the programming language and common libraries. Basic rules for portable code are defined by \person{Kerighan} and \person{Pike} \cite{kernighan99}. Portable code lets software spread faster.
meillo@189 92
meillo@189 93
meillo@189 94 \subsubsection*{Usability}
meillo@196 95 Usability, not mentioned by \person{Hafiz} (he focuses on architecture) but by \person{Spinellis} and \person{Kan}, is a property very important from the user's point of view. Software with bad usability is rarely used, no matter how good it is. If roughly equivilent substitutes with better usability exist, the user will switch to one of them. Usability here means easy to set up and configure, too; users are also administrators here. Having \mta{}s on many home servers and clients, requires easy and standardized configuration. The common setups should be configurable with single actions by the user. Complex configuration should be possible, but focused must be the most common form of configuration: choosing one of several standard setups.
meillo@146 96
meillo@146 97
meillo@146 98
meillo@146 99
meillo@185 100
meillo@185 101 \subsection{Functional requirements}
meillo@185 102
meillo@185 103 This section identifies the needed functionality for a modern \MTA. The basic job of a \mta\ is to tranport mail from a sender to a recipient. This is the definition of such kind of software and this is how \MTA{}s are generally seen \cite[page 19]{dent04} \cite[pages 3-5]{hafiz05}.
meillo@185 104
meillo@185 105 An \MTA\ therefore needs at least a mail receiving facility and a mail sending facility.
meillo@185 106
meillo@185 107
meillo@185 108
meillo@185 109 \subsubsection*{Incoming channels}
meillo@185 110
meillo@185 111 \sendmail-compatible \mta{}s must support at least two incoming channels: mail submitted using the \sendmail\ command, and mail received via the \SMTP\ daemon. Thus it is common to split the incoming channel into local and remote. This is done by \qmail\ and \postfix. The same way is \person{Hafiz}'s view. \SMTP\ is the primary mail transport protocol today, but with the increasing need for new protocols\ref{FIXME} in mind, support for more than just \SMTP\ is good to have. This leads to more than one remote channel.
meillo@185 112
meillo@185 113
meillo@185 114 \subsubsection*{Outgoing channels}
meillo@185 115
meillo@185 116 Outgoing mail is commonly either sent using \SMTP, piped into local commands (for example \texttt{uucp}), or delivered locally by appending to a mailbox.
meillo@185 117
meillo@185 118 Outgoing channels are similar for \qmail, \postfix, and \name{sendmail X}: All of them have a module to send mail using \SMTP, and one for writing into a local mailbox. Local mail delivery is a job that requires root priveledge to be able to switch to any user in order to write to his mailbox.
meillo@185 119
meillo@185 120 As mail delivery to local users, is \emph{not} included in the basic job of an \MTA{}, why should it care about it? In order to keep the system simple and to have programs that do one job well, the local delivery job should be handed over to a specialist: the \name{mail delivery agent}. \NAME{MDA}s know about the various mailbox formats and are aware of the problems of concurrent write access and thelike. Hence handling the message and the responsiblity over to a \NAME{MDA}, like \name{procmail} or \name{maildrop}, seems to be the right way to go.
meillo@185 121
meillo@185 122 This means an outgoing connection that pipes mail into local commands is required. Other outgoing channels, one for each supportet protocol, may be designed like it was done in other \MTA{}s.
meillo@185 123
meillo@185 124
meillo@185 125
meillo@185 126 \subsubsection*{Mail queue}
meillo@185 127
meillo@185 128 Additionally to the mail receiving and sending facilities, mail queues are a basic feature. A mail queue removes the need to deliver intantly as a message is received. They provide fail-safe storage of mails until they are delivered. Mail queues are probably used in all \mta{}s, excluding the simple forwarders. A mail queue is a essential requirement for \masqmail, as it is to be used for non-permanent online connections. This means, mail must be queued until a online connection is available to send the message.
meillo@185 129
meillo@185 130 The mail queue and the module to manage it are the central part of the whole system. This demands especially for robustness and reliability, as a failure here can lead to loosing mail. An \MTA\ takes over responsibility for mail in accepting it, hence loosing mail messages is absolutely to avoid. This covers any kind of crash situation too. The worst thing acceptable to happen is a mail to be sent twice.
meillo@185 131
meillo@185 132 \sendmail, \exim, \qmail, \name{sendmail X}, and \masqmail\ feature one single mail queue. \postfix\ has three of them: \name{incoming}, \name{active}, and \name{deferred}. (The \name{maildrop} queue is excluded, as it is only used for the \texttt{sendmail} command.)
meillo@185 133
meillo@185 134 \MTA\ setups that do external content scanning tend to require two separate queues. To use \sendmail\ in such setups requires two independent instances, with two separate queues, running. \exim\ can handle it with special \name{router} and \name{transport} rules, but the data flow gets complicated. Having two independent queues seems to be preferable.
meillo@185 135
meillo@185 136
meillo@185 137
meillo@185 138
meillo@185 139 \subsubsection*{Header sanitizing}
meillo@185 140
meillo@185 141 Mail coming into the system often lacks important header lines. At least the required ones must be added from the \MTA. One example is the \texttt{Date:} header, another is the, not required but recommended \texttt{Message-ID:} header. Apart from adding missing headers, rewriting headers is important too, to change the locally known domain part of email addresses to globally known ones for example. \masqmail\ needs also the ability to rewrite the domain part dependent on the route used to send the message.
meillo@185 142
meillo@185 143 Generating the envelope is a related job. The envelope specifies the actual recipient of the mail, no matter what the \texttt{To:}, \texttt{Cc:}, and \texttt{Bcc:} headers tell. Multiple reciptients lead to multiple different envelopes, containing all the same mail message.
meillo@185 144
meillo@185 145
meillo@185 146
meillo@185 147
meillo@185 148 \subsubsection*{Aliasing}
meillo@185 149
meillo@185 150 Email addresses can have aliases and need to be expanded. Aliases can be of different kind: different local user, remote user, list of local and/or remote users, or a command. Most important are the aliases in the \path{aliases} file, usually located at \path{/etc/aliases}. Addresses expanding to lists of users lead to more envelopes. Aliases changing the reciptients domain part may require a different route to use.
meillo@185 151
meillo@185 152
meillo@185 153
meillo@185 154
meillo@185 155 \subsubsection*{Choose route to use}
meillo@185 156
meillo@185 157 One key feature of \masqmail\ is its ability to send mail out in different ways. The decision is based on the current online state and whether a route may be used for a message or not. The online state can be retrieved in tree ways, explained in \ref{sec:fixme}. A route to send is found by checking every available route for being able to transfer the current message, until one matches.
meillo@185 158
meillo@185 159
meillo@185 160
meillo@185 161
meillo@185 162 \subsubsection*{Authentication}
meillo@185 163
meillo@185 164 One thing to avoid is being an \name{open relay}. Open relays allow to relay mail from everywhere to everywhere. This is a major source of spam. The solution is restricting relay\footnote{Relaying is passing mail, that is not from and not for the own system, through it.} access.
meillo@185 165
meillo@185 166 Several ways to restrict access are available. The most simple one is restrictiction relaying by the \NAME{IP} address. No extra complexity is added this way, but static \NAME{IP} addresses are needed.
meillo@185 167
meillo@185 168 If static access restriction is not possible, for example mail from locations with changing \NAME{IP} addresses, some kind of authentication mechanism is required. Three common kinds exist:
meillo@185 169 \begin{itemize}
meillo@185 170 \item \SMTP-after-\NAME{POP}: Uses authenication on the \NAME{POP} protocol to permit incoming \SMTP\ connections for a limited time afterwards.
meillo@185 171 \item \SMTP authentication: An extension to \SMTP. Authentication can be requested before mail is accepted.
meillo@185 172 \item Certificates: They confirm the identity of someone.
meillo@185 173 \end{itemize}
meillo@185 174
meillo@185 175
meillo@185 176
meillo@185 177 \subsubsection*{Encryption}
meillo@185 178
meillo@185 179 Electronic mail is very weak to sniffing attacks, because all data transfer is unencrypted. This concerns the message's content, as well as the email addresses in header and envelope, but also authentication dialogs that may transfer plain text passwords (\NAME{PLAIN} and \NAME{LOGIN} are examples). Thus encryption is wanted.
meillo@185 180
meillo@185 181 The common way to encrypt \SMTP\ dialogs is using \name{Transport Layer Security} (short: \TLS, successor of \NAME{SSL}). \TLS\ encrypts the datagrams of the \name{transport layer}. This means it works below the application protocols and can be used by any of them\citeweb{wikipedia:tls}.
meillo@185 182
meillo@185 183 Outgoing \SMTP\ connections can get encrypted using a secure tunnel, created by an external application. Incoming connections, can not use this technique because the remote \NAME{IP} address is hidden then; \NAME{STARTTLS}---defined in \RFC2487---is what \mta{}s implement.
meillo@185 184
meillo@185 185
meillo@185 186
meillo@185 187 \subsubsection*{Spam prevention}
meillo@185 188
meillo@185 189 Spam is a major threat nowadays, but it is a war in which the good guys tend to lose. Putting much effort in fighting spam results in few gain. Real success will only be possible with new---better---protocols and abandonning the weak legacy technologies. The goal is to provide state-of-the-art spam protection, but not more (see section \ref{sec:swot-analysis}). As spam is not just a nuisance for end users, but also for the infrastructure---the \mta{}s---by increasing the amount of mail messages, \MTA{}s need to protect themself.
meillo@185 190
meillo@185 191 Filtering spam can be done in two ways: Refusing spam during the \SMTP\ dialog or checking for spam after the mail was accepted and queued. Both have advantages and disadvantages, so modern \MTA{}s use them in combination. Spam is identified by the results of a set of checks. Static rules, querying databases (\NAME{DNS} blacklists), requesting special client behavior (\name{greylisting}, \name{hashcash}), or statistical analysis (\name{bogofilter}) are checks that may be used. Running more checks leads to better results, but takes more system resources and time.
meillo@185 192
meillo@185 193 Doing some basic checks during the \SMTP\ dialog seems to be a must.
meillo@185 194
meillo@185 195
meillo@185 196
meillo@185 197 \subsubsection*{Virus checking}
meillo@185 198
meillo@185 199 Related to spam is malicous content (short: \name{malware}) like viruses, worms, trojan horses. They, in contrast to spam, do not affect the \MTA\ itself, as they are in the mail body. \MTA{}s searching for malware is equal to real world's post offices opening letters to check if they contain something that could harm the recipient. This is not a mail transport job. But the last \MTA---the one responsible for the recipient---seems to be at a good position to do this work.
meillo@185 200
meillo@185 201 In any way should malware checking be done by external programs that may be invoked by the \mta. But using mail deliver and processing agents, like \name{procmail}, a better suited locations to invoke content scanners.
meillo@185 202
meillo@185 203 A popular email filter framework is \name{amavis} which integrates various spam and virus scanners. The common setup includes a receiving \MTA\ which sends it to \name{amavis} using \SMTP, \name{amavis} processes the mail and sends it then to a second \MTA\ that does the outgoing transfer. Having interfaces to such scanners is, for sure, good to have.
meillo@185 204
meillo@185 205
meillo@185 206
meillo@185 207 \subsubsection*{Archiving}
meillo@185 208
meillo@185 209 Mail archiving and auditability become more important as electronic mail becomes more important. The ability to archive verbatim copies of every mail coming into and every mail going out of the system, with relation between them, appears to be a goal to achieve.
meillo@185 210
meillo@185 211 \postfix\ for example has a \texttt{always\_bcc} feature, to send a copy of every mail to a definable reciptient. At least this funtionality should be given, although a more complete approach is preferable.
meillo@185 212
meillo@185 213
meillo@185 214
meillo@185 215
meillo@185 216
meillo@185 217
meillo@185 218
meillo@185 219
meillo@185 220 \section{Work to do}
meillo@185 221
meillo@187 222 After \masqmail's features were presented in section \ref{sec:fixme} and the requirements for modern \mta{}s were identified in section \ref{sec:fixme}, here the differences between them are shown.
meillo@185 223
meillo@185 224
meillo@187 225
meillo@187 226 \subsubsection*{Fulfilled requirements}
meillo@187 227
meillo@187 228 \masqmail's incoming and outgoing channels are the common ones: the \texttt{sendmail} command and \SMTP\ for incoming mail; local delivery, piping to commands, and \SMTP\ for outgoing mail. Support for other protocols is not available. To add it, modifications at many places in the source are needed.
meillo@187 229
meillo@187 230 One single mail queue is used in \masqmail. The envelope and mail headers are generated when the mail is put into the queue. Aliasing is done on delivery, after the route to be used was determined. Headers may be rewritten then. These parts do all provide the functionality required.
meillo@187 231
meillo@187 232 Static authentication, based on \NAME{IP} addresses, can be set up using the \path{hosts.allow} and \path{hosts.deny} files. Dynamic authentication is supported in form of \NAME{SMTP-AUTH} and \SMTP-after-\NAME{POP}, but only for outgoing connections. The same for encryption which is also only available for outgoing \SMTP\ connections; here a wrapper application like \name{openssl} needs to be used. Support for authentication and encryption of incoming connections is completely missing, but a basic requirement for all secure emailing.
meillo@187 233
meillo@187 234 \masqmail\ does not provide special support for spam filter or content checking. But it is possible to invoke external filter applications by running two independent instances of \masqmail, connected by the filter application. The receiving \MTA\ instance accepts mail and pushes it into the filter. The filter application receives mail, processes it, possible modifies it, and pushes it over to a second \MTA\ instance. The second \MTA\ is responsible for further delivery of the mail. Appendix \ref{app:FIXME} shows configuration files to create such a setup. This is a concept that works in general. However, real spam \emph{prevention}---to not accept spam mail at all---or good filter interfaces are not available, but are nessesary for using \masqmail\ in an unsafe environment.
meillo@187 235
meillo@187 236 There is currently no way of archiving every message going through \masqmail.
meillo@187 237
meillo@187 238
meillo@196 239 Non-functional requirements are not so easy to be marked as fulfilled or not. Instead they are discussed here.
meillo@196 240
meillo@196 241 \masqmail\ needs to be ``secure enough'', but what is ``secure enough''? This depends on its target field. Currently \masqmail\ is targeted to workstations and private networks, with explicit warning for use on permanent online hosts \citeweb{masqmail:homepage2}. \masqmail's current security is bad. (For instance does a long time known attack against \sendmail, described by \person{Sill} \cite[page~4]{sill02}, still outwit \masqmail). The security, however, seems acceptable for use on workstations and private networks, if the environment is trusted. In environments where untrusted components or persons have access to \masqmail, its security is too low.
meillo@196 242
meillo@196 243 Similar for its reliability. It has been reported that \masqmail\ has not sent mail under some circumstances \citeweb{FIXME}. %fixme
meillo@196 244 The author also noticed problems with where only one part of the queued message was removed, the other part remained as garbage. Fortunately, reports abou lost mail are not known. Current reliability is not good enough.
meillo@196 245
meillo@196 246 The logging behavior of \masqmail\ is good, but does not cover everything. Also it trusts its environment. For example, if the queue directory is world writeable by accident (or as action of an intruder), any user can remove messages from the queue or replace them with own ones. \masqmail\ does not even write a debug message in this case.
meillo@196 247
meillo@196 248 \masqmail's extendability is very poor. This is a general problem of monolithic software, but can thus be provided with high effort. \exim\ is an example for good extendability for a monolithic program.
meillo@196 249
meillo@196 250 The maintainability of \masqmail\ appears to be equivilent to other software of similar kind. Missing modularity and therefore more complexity makes the maintainer's work harder. In summary is \masqmail's maintainability bearable, and in view of its missing modularity good. The testability suffers from the same. Testing program parts is hard, but done by compiling source parts to special test programs.
meillo@196 251
meillo@196 252 The performance---effenciency---of \masqmail\ is good enough for its target field of operation, where this is a minor goal. As well is its availability. Hence no further work needs to be done her.
meillo@196 253
meillo@196 254 The code's protability is good with view on \unix-like operation systems. At least \name{Debian}, \name{Redhat}, \NAME{SUSE}, \name{slackware}, \name{Free}\NAME{BSD}, \name{Open}\NAME{BSD}, and \name{Net}\NAME{BSD} are reported to be able to compile and run \masqmail\ \citeweb{masqmail:homepage2}. About special requirements for the underlying file system is not heard of. The portability is already good.
meillo@196 255
meillo@196 256 The usability, from the administator's point of view, is very good. \masqmail\ was developed to suite a specific, limited job, its configuration does perfect match. The user's view does not reach to the \MTA, as it is hidden behind the \name{mail user agent}.
meillo@196 257
meillo@187 258
meillo@187 259
meillo@187 260 \subsubsection*{Missing parts}
meillo@187 261
meillo@187 262 Support for other protocols than \SMTP\ seems not to be nessesary at the moment. Adding such support will need lots of work in all parts of \masqmail, hence delaying it until the support is needed appears to be the best solution.
meillo@187 263
meillo@187 264 Authentication of incoming \SMTP\ connections is definately needed and should be added soon. The same applies to encryption of incomming connections. These two features are essential for restricting relaying and providing privacy.
meillo@187 265
meillo@187 266 As authentication can be a guard against spam, filter facilities have lower priority. But basic spam filtering and interfaces for external tools should be implemented in future. Content checking, if really nessesary, should be left over to the \NAME{MDA}, to deal with it in local delivery.
meillo@187 267
meillo@187 268 Archiving again is prefered to be implemented soon. It does not require much work, but enables all kinds of statistical analysis. Also it is a requirement for companies to archive their mail communication.
meillo@187 269
meillo@196 270 Non-functional requirements need improvements too.
meillo@185 271
meillo@196 272 \masqmail's security is bad and limits it thus to a field of operation that shrinks. Security becomes more important and networking and interaction increases. Save and trusted environment become rare. Improving security is an important thing to do in future.
meillo@185 273
meillo@196 274 Reliability is also to improve. It is a key quality property for an \MTA, and not good enough in \masqmail. Also is the program lacking robustness. Checking the environment and reporting bad characteristics is wanted. Especially improving robustness inrelation to the queue is favorable; applying ideas of \name{crash-only software}\cite{candea03} will be a good step.
meillo@185 275
meillo@196 276 Extendability, maintainability, and testability do all lack from the monolithic architecture and are nearly impossible to improve without changing the programs structure. These properties can hardly be retrofitted into software. But extendability might become important in the future, and the other two make all further work on the software easier.
meillo@185 277
meillo@196 278 Performance is a nice-to-have property, but as performance improvements are in contrast to many other quality poperties (reliability, maintainability, usability, capability \cite[page~5]{kan03}), jeopardizing these to gain some more performance should not be done. \person{Kernighan} and \person{Pike} state clear: ``[T]he first principle of optimization is \emph{don't}.''\cite[page~165]{kernighan99}. \masqmail\ is not a program to be used on large servers, but on small devices. Thus important for \masqmail\ is focusing on energy and heat, maybe also system resources, not on performance.
meillo@196 279
meillo@196 280 Focusing on being portable amoung the various flavors of \unix\ systems seems to be okay, because these systems are the ones \MTA{}s run on usually. Problems with non-\unix platforms are primary expected to come from file systems lacking required features. But no special care should be taken here.
meillo@196 281
meillo@196 282 Configuration could be eased more, by providing configuration generators to be able to run \masqmail\ right ``out of the box'' with only running one of several configuration scripts for common setups. This would make \masqmail\ better usable by people not technical educated.
meillo@196 283
meillo@196 284
meillo@196 285
meillo@196 286
meillo@196 287
meillo@196 288
meillo@196 289 \section{Discussion on architecture}
meillo@146 290
meillo@187 291 %fixme: why is there a need for a new arch??
meillo@188 292 Adding authentication and encryption support is limited to a narrow region in the code. Such features are addable to the current code base without much problem. In contrast do support for new protocols or mail processing interfaces to external programs require a lot of effort. Changes in many parts of the source code are required. It is no good idea to implement large retro-fitted features in a software that is critical in security and reliability, like \masqmail. Worse if these features need changes in the program's structure, like adding mail scanning interfaces would do.
meillo@187 293
meillo@188 294 If such features are needed, it is best do redesign the program's structure and rebuild it. A program's structure is primary its architecture. Which is probably the most influencing design decision, and has the greatest impact on the program's future capabilities. The architecture defines what the program can do, how it can be used. If the architecture does not fit the requirements, develpement reached a dead end \dots\ further work will make everything worse. The only good solution is to change the architecture, which, sadley but most likely, means a redesign from scratch.
meillo@187 295
meillo@188 296 \masqmail's current artitecture is monolitic like \sendmail's and \exim's. But more than the other two, is it one block of interweaved code. \sendmail\ provides now, with its \name{milter} interface, standardized connection channels to external modules. \exim\ has a highly structured code with many internal interfaces, a good example is the one for authentication ``modules''. %fixme: add ref
meillo@188 297 \masqmail\ has none of them; it is what \sendmail\ was in the beginning: a single large block.
meillo@161 298
meillo@161 299 Figure \ref{fig:masqmail-arch} is an attempt to depict \masqmail's internal structure.
meillo@161 300
meillo@161 301 \begin{figure}
meillo@161 302 \begin{center}
meillo@161 303 \input{input/masqmail-arch.tex}
meillo@161 304 \end{center}
meillo@188 305 \caption{Internal structure of \masqmail}
meillo@161 306 \label{fig:masqmail-arch}
meillo@161 307 \end{figure}
meillo@161 308
meillo@163 309 \sendmail\ improved its old architecture, for example by adding the milter interface. \exim\ was designed and is carefully maintained with a modular-like code structure in mind. \qmail\ started from scratch with a ``security-first'' approach, \postfix\ improved on it, and \name{sendmail X}/\name{MeTA1} tries to adopt the best of \qmail\ and \postfix, to completely replace the old \sendmail\ architecture. \person{Hafiz} \cite{hafiz05}. describes this evolution of \mta\ architecture very well.
meillo@161 310
meillo@188 311 Every one of the popular \MTA{}s is more modular, or became more modular over time, than \masqmail\ is. Modern requirements like spam protection and future requirements like---probably---the use of new mail transport protocols demand modular designs for keeping the software simple. Simplicity is a key property for security. ``[T]he essence of security engenieering is to build systems that are as simple as possible.''\cite[page 45]{graff03}
meillo@161 312
meillo@188 313 \person{Hafiz} agrees: ``The goal of making software secure can be better achieved by making the design simple and easier to understand and verify.''\cite[page 64]{hafiz05} He identifies the security of \qmail\ to come from it's \name{compartmentalization}, which goes hand in hand with modularity:
meillo@163 314 \begin{quote}
meillo@188 315 A perfect example is the contrast between the feature envy early \sendmail\ architecture implemented as one process and the simple, modular architecture of \qmail. The security of \qmail\ comes from its compartmentalized simple processes that perform one task only and are therefore testable for security. \cite[page 64]{hafiz05}
meillo@163 316 \end{quote}
meillo@188 317 As well does \person{Dent}: ``The modular architecture of Postfix forms the basis for much of its security.''\cite[page 7]{dent04} Modularity is also needed to satisfy modern \MTA\ requirements, in providing a clear interface to add functionality without increasing the overall complexity much.
meillo@161 318
meillo@188 319 Security comes from good design, as \person{Graff} and \person{van Wyk} explain:
meillo@163 320 \begin{quote}
meillo@163 321 Good design is the sword and shield of the security-conscious developer. Sound design defends your application from subversion or misuse, protecting your network and the information on it from internal and external attacks alike. It also provides a safe foundation for future extensions and maintainance of the software.
meillo@163 322 %
meillo@163 323 %Bad design makes life easier for attackers and harder for the good guys, especially if it contributes to a false sends of security while obscuring pertinent failings.
meillo@163 324 \cite[page 55]{graff03}
meillo@163 325 \end{quote}
meillo@161 326
meillo@161 327
meillo@188 328 All this leads to a rewrite of \masqmail, using a modern, modular architecture, \emph{if} further features need to be added, ones that require changes in \masqmail's structure. As well is a rewrite needed, if \masqmail\ should become a modern \MTA, with good quality properties.
meillo@161 329
meillo@188 330 But redesigning and rewriting a software from scratch is hard. It takes time to design a new architecture, which must prove it is secure and reliable. And much time and work is needed to implement the design, test it, fix bugs, and so on. Thus the gain of a new design must overweight the effort needed to spend.
meillo@161 331
meillo@188 332 \person{Wheeler}'s program \name{sloccount} calculates following estimations for \masqmail's code base as of version 0.2.21 (excluding library code):
meillo@188 333 {\small
meillo@188 334 \begin{verbatim}
meillo@188 335 Total Physical Source Lines of Code (SLOC) = 9,041
meillo@188 336 Development Effort Estimate, Person-Years (Person-Months) = 2.02 (24.22)
meillo@188 337 (Basic COCOMO model, Person-Months = 2.4 * (KSLOC**1.05))
meillo@188 338 Schedule Estimate, Years (Months) = 0.70 (8.39)
meillo@188 339 (Basic COCOMO model, Months = 2.5 * (person-months**0.38))
meillo@188 340 Estimated Average Number of Developers (Effort/Schedule) = 2.89
meillo@188 341 Total Estimated Cost to Develop = $ 272,690
meillo@188 342 (average salary = $56,286/year, overhead = 2.40).
meillo@188 343 SLOCCount, Copyright (C) 2001-2004 David A. Wheeler
meillo@188 344 \end{verbatim}
meillo@188 345 }
meillo@188 346 The development cost is not relevant for a \freesw\ project of volunteer developers, but the time needed is. About 24 man-months are estimated. The current code base was written almost completely by Oliver \person{Kurth} within four years, in his spare time. This means he needed around twice as much time.
meillo@161 347
meillo@188 348 Given the asumptions that an equal amount of code is to be produced, a third of existing code can be reused plus concepts and experience, and development speed is like \person{Kurth}'s. Then it would take about two years to have a redesigned new \masqmail\ with the same features. Less time could be needed if a simpler architecture allows faster develpement, better testing, and less bugs.
meillo@177 349
meillo@188 350 The further section describes a new modern design for \masqmail. A plan to rewrite \masqmail\ from scratch.
meillo@177 351
meillo@177 352
meillo@177 353
meillo@188 354 %\person{Hafiz} adds: ``The major idea is that security cannot be retrofitted into an architecture.''\cite[page 64]{hafiz05}
meillo@177 355
meillo@177 356
meillo@177 357
meillo@177 358
meillo@196 359
meillo@196 360
meillo@196 361
meillo@196 362
meillo@196 363
meillo@196 364 \section{Result}
meillo@196 365
meillo@196 366 Directions to go
meillo@196 367
meillo@196 368 Now how could \masqmail\ be like in, say, five years?
meillo@196 369
meillo@196 370 This section discusses about what shapes \masqmail\ could have---which directions the development could go to.
meillo@196 371
meillo@196 372
meillo@196 373
meillo@196 374
meillo@196 375 1) fix the current version
meillo@196 376
meillo@196 377
meillo@196 378 2) create a new one
meillo@196 379 But how is the effort of this complete rewrite compared to what is gained afterwards?
meillo@196 380 << would one create it at all? >>
meillo@196 381
meillo@196 382
meillo@196 383 pro---contra
meillo@196 384
meillo@196 385
meillo@196 386
meillo@196 387
meillo@196 388 << short term goals --- long term goals >>
meillo@196 389
meillo@196 390 do it like sendmail: first do the most needed stuff on the old design to make it still usable. Then design a new version from scratch, for the future.
meillo@196 391
meillo@196 392 << which parts to take out and do within the thesis >>
meillo@196 393
meillo@196 394
meillo@196 395
meillo@196 396
meillo@196 397
meillo@196 398
meillo@196 399
meillo@196 400
meillo@196 401
meillo@196 402
meillo@196 403
meillo@196 404
meillo@196 405 \chapter{A design from scratch}
meillo@177 406
meillo@185 407 The last sections identified the jobs that need to be done by a modern \MTA; problems and prefered choices were mentioned too. Now the various jobs are assigned to modules, of which an architecture is created. It is inpired by existing ones and driven by the identified jobs and requirements.
meillo@161 408
meillo@163 409
meillo@163 410
meillo@163 411
meillo@185 412 \subsection{Design decisions}
meillo@185 413
meillo@185 414 One major design idea of the design were:
meillo@185 415 \begin{itemize}
meillo@185 416 \item free the internal system from in and out channels
meillo@185 417 \item arbitrary protocol handlers have to be addable afterwards
meillo@185 418 \item a single facility for scanning (all mail goes through it)
meillo@185 419 \item concentrate on mail transfer
meillo@185 420 \end{itemize}
meillo@163 421
meillo@163 422
meillo@163 423 \subsubsection*{Incoming channels}
meillo@163 424
meillo@170 425 \sendmail-compatible \mta{}s must support at least two incoming channels: mail submitted using the \sendmail\ command, and mail received via the \SMTP\ daemon. It is therefor common to split the incoming channel into local and remote. This is done by \qmail\ and \postfix. The same way is \person{Hafiz}'s view.
meillo@163 426
meillo@170 427 In contrast is \name{sendmail X}: Its locally submitted messages go to the \SMTP\ daemon, which is the only connection towards the mail queue. %fixme: is it a smtp dialog? or a second door?
meillo@170 428 \person{fanf} proposes a similar approach. He wants the \texttt{sendmail} command to be a simple \SMTP\ client that contacts the \SMTP\ daemon of the \MTA\ like it is done by connections from remote. The advantage here is one single module where all \SMTP\ dialog with submitters is done. Hence one single point to accept or refuse incoming mail. Additionally does the module to put mail into the queue not need to be \name{setuid} or \name{setgid} because it is only invoked from the \SMTP\ daemon. The \MTA's architecture would become simpler and common tasks are not duplicated in modules that do similar jobs.
meillo@163 429
meillo@170 430 But merging the input channels in the \SMTP\ daemon makes the \MTA\ heavily dependent on \SMTP\ being the main mail transfer protocol. To \qmail\ and \postfix\ new modules to support other ways of message receival may be added without change of other parts of the system. Also is it better to have more independent modules if each one is simpler then.
meillo@163 431
meillo@170 432 With the increasing need for new protocols in mind, it seems better to have single modules for each incoming channel, although this leads to duplicated acceptance checks.
meillo@163 433
meillo@163 434
meillo@163 435 \subsubsection*{Outgoing channels}
meillo@163 436
meillo@170 437 Outgoing mail is commonly either sent using \SMTP, piped into local commands (for example \texttt{uucp}), or delivered locally by appending to a mailbox.
meillo@170 438
meillo@163 439 Outgoing channels are similar for \qmail, \postfix, and \name{sendmail X}: All of them have a module to send mail using \SMTP, and one for writing into a local mailbox. Local mail delivery is a job that requires root priveledge to be able to switch to any user in order to write to his mailbox. Modular \MTA{}s do not need \name{setuid root}, but the local delivery process (or its parent) needs to run as root.
meillo@163 440
meillo@170 441 As mail delivery to local users, is \emph{not} included in the basic job of an \MTA{}, why should it care about it? In order to keep the system simple and to have programs that do one job well, the local delivery job should be handed over to a specialist: the \name{mail delivery agent}. \NAME{MDA}s know about the various mailbox formats and are aware of the problems of concurrent write access and thelike. Hence handling the message and the responsiblity over to a \NAME{MDA}, like \name{procmail} or \name{maildrop}, seems to be the right way to go.
meillo@163 442
meillo@170 443 This means an outgoing connection that pipes mail into local commands is required. Other outgoing channels, one for each supportet protocol, may be designed like it was done in other \MTA{}s.
meillo@170 444
meillo@170 445
meillo@170 446
meillo@170 447 \subsubsection*{Mail queue}
meillo@170 448
meillo@170 449 Mail queues are probably used in all \mta{}s, excluding the simple forwarders. A mail queue is a essential requirement for \masqmail, as it is to be used for non-permanent online connections. This means, mail must be queued until a online connection is available to send the message.
meillo@170 450
meillo@170 451 The mail queue and the module to manage it are the central part of the whole system. This demands especially for robustness and reliability, as a failure here can lead to loosing mail. An \MTA\ takes over responsibility for mail in accepting it, hence loosing mail messages is absolutely to avoid. This covers any kind of crash situation too. The worst thing acceptable to happen is a mail to be sent twice.
meillo@170 452
meillo@170 453 \sendmail, \exim, \qmail, \name{sendmail X}, and \masqmail\ feature one single mail queue. \postfix\ has three of them: \name{incoming}, \name{active}, and \name{deferred}. (The \name{maildrop} queue is excluded, as it is only used for the \texttt{sendmail} command.)
meillo@170 454
meillo@170 455 \MTA\ setups that include content scanning tend to require two separate queues. To use \sendmail\ in such setups requires two independent instances, with two separate queues, running. \exim\ can handle it with special \name{router} and \name{transport} rules, but the data flow gets complicated. Hence an idea is to use two queues, \name{incoming} and \name{active} in \postfix's terminology, with the content scanning within the move from \name{incoming} to \name{active}.
meillo@163 456
meillo@182 457 \sendmail, \exim, \qmail, and \masqmail\ all use at least two files to store one message in the queue: one file contains the message body, another the envelope and header information. The one containing the mail body is not modified at all. \postfix\ takes a different approach in storing queued messages in an internal format within one file. \person{Finch} takes yet another different approach in suggesting to store the whole queue in one single file with pointers to separating positions \cite{finchFIXME}.
meillo@182 458 %fixme: check, cite, and think about
meillo@182 459
meillo@163 460
meillo@165 461
meillo@165 462 \subsubsection*{Sanitize mail}
meillo@165 463
meillo@170 464 Mail coming into the system often lacks important header lines. At least the required ones must be added from the \MTA. A good example is the \texttt{Message-Id:} header.
meillo@165 465
meillo@170 466 In \postfix, this is done by the \name{cleanup} module, which invokes \name{rewrite}. The position in the message flow is after coming from one of the several incoming channels and before the message is stored into the \name{incoming} queue. Modules that handle incoming channels may also add headers, for example the \texttt{From:} and \texttt{Date:} headers. \name{cleanup}, however, does a complete check to make the mail header complete and valid.
meillo@170 467
meillo@170 468 Apart from deciding where to sanitize the mail header, is the question where to generate the envelope. The envelope specifies the actual recipient of the mail, no matter what the \texttt{To:}, \texttt{Cc:}, and \texttt{Bcc:} headers tell. Multiple reciptients lead to multiple different envelopes, containing all the same mail message.
meillo@170 469
meillo@170 470
meillo@170 471
meillo@165 472 \subsubsection*{Aliasing}
meillo@165 473
meillo@170 474 Where should aliases get expanded? They appear in different kind. Important are the ones available in the \path{aliases} file. Aliases can be:
meillo@170 475 \begin{itemize}
meillo@170 476 \item a different local user (e.g.\ ``\texttt{bob: alice}'')
meillo@170 477 \item a remote user (e.g.\ ``\texttt{bob: john@example.com}'')
meillo@170 478 \item a list of users (e.g.\ ``\texttt{bob: alice, john@example.com}'')
meillo@170 479 \item a command (e.g.\ ``\texttt{bob: |foo}'')
meillo@170 480 \end{itemize}
meillo@170 481 Addresses expanding to lists of users lead to more envelopes. Aliases changing the reciptients domain part may require a different route to use.
meillo@165 482
meillo@172 483 Aliasing is often handled in expanding the alias and reinjecting the mail into the system. Unfortunately, the mail is processed twice then; additionally does the system have to handle more mail this way. If it is wanted to check the new recipient address for acceptance and do all processing again, then reinjecting it is the best choice.
meillo@163 484
meillo@163 485
meillo@163 486
meillo@185 487 \subsubsection*{Choose route to use}
meillo@185 488
meillo@185 489 One key feature of \masqmail\ is its ability to send mail out in different ways. The decision is based on the current online state and whether a route may be used for a message or not. The online state can be retrieved in tree ways, explained in \ref{sec:fixme}. A route to send is found by checking every available route for being able to transfer the current message, until one matches.
meillo@185 490
meillo@185 491 This functionality should be implemented in the module that is responsible to invoke one of the outgoing channel modules (for example the one for \SMTP\ or the pipe module).
meillo@185 492
meillo@185 493 \masqmail\ can rewrite the envelope's from address and the \texttt{From:} header, dependent on the outgoing route to use. This rewrite must be done \emph{after} it is clear which route a mail will take, of course, so this may be not the module where other header editing is done.
meillo@185 494 %fixme: see hafiz05 page 57: maybe put the rewriting into the sending module (like smx, exim, courier) (problem with archiving of all outgoing mail?)
meillo@185 495
meillo@185 496
meillo@185 497
meillo@163 498 \subsubsection*{Authentication}
meillo@163 499
meillo@170 500 One thing to avoid is being an \name{open relay}. Open relays allow to relay mail from everywhere to everywhere. This is a major source of spam. The solution is restricting relay\footnote{Relaying is passing mail, that is not from and not for the own system, through it.} access.
meillo@163 501
meillo@170 502 Several ways to restrict access are available. The most simple one is restrictiction by the \NAME{IP} address. No extra complexity is added this way, but static \NAME{IP} addresses are mandatory. This kind of restriction may be enabled using the operating system's \path{hosts.allow} and \path{hosts.deny} files. To allow only connections to port 25 from localhost or the local network \texttt{192.168.100.0/24} insert the line ``\texttt{25: ALL}'' into \path{hosts.deny} and ``\texttt{25: 127.0.0.1, 192.168.100.}'' into \path{hosts.allow}.
meillo@170 503
meillo@170 504 If static access restriction is not possible, for example if mail from locations with changing \NAME{IP} addresses wants to be accepted, some kind of authentication mechanism is required. Three common kinds exist:
meillo@170 505 \begin{enumerate}
meillo@185 506 \item \SMTP-after-\NAME{POP}: uses authenication on the \NAME{POP} protocol to permit incoming \SMTP\ connections for a limited time afterwards.
meillo@185 507 \item \SMTP authentication: is an extension to \SMTP. Authentication can be requested before mail is accepted.
meillo@185 508 \item Certificates: confirm the identity of someone.
meillo@170 509 \end{enumerate}
meillo@170 510
meillo@170 511
meillo@163 512
meillo@163 513 \subsubsection*{Encryption}
meillo@163 514
meillo@172 515 Electronic mail is very weak to sniffing attacks, because all data transfer is unencrypted. This concerns the message's content, as well as the email addresses in header and envelope, but also authentication dialogs that may transfer plain text passwords (\NAME{PLAIN} and \NAME{LOGIN} are examples). Adding encryption is therefor wanted.
meillo@165 516
meillo@172 517 The common way to encrypt \SMTP\ dialogs is using \name{Transport Layer Security} (short: \TLS, successor of \NAME{SSL}). \TLS\ encrypts the datagrams of the \name{transport layer}. This means it works below the application protocols and can be used by any of them\citeweb{wikipedia:tls}.
meillo@165 518
meillo@172 519 \TLS\ allows to create secure tunnels through which arbitrary programs can communicate. Hence one can add secure communication afterwards to programs without changing them. \name{OpenSSL} for example---a free implementation---allows traffic to be piped into a command; a secure tunnel is created and the traffic is forwarded through it. Or a secure tunnel can be set up between a local and a remote port; this tunnel can then be used by any application.
meillo@165 520
meillo@172 521 The \NAME{POP} protocol, for example, is good suited for such tunneling, but \SMTP\ is is not generally. Outgoing \SMTP\ client connections can be tunneled without problem---\masqmail\ already provides a configure option called \texttt{wrapper} to do so. Tunneling incomming connections to a server leads to problems with \SMTP. As data comes encrypted through the tunnel to the receiving host and gets then decrypted and forwarded on local to the port the application listens on. From the \MTA's view, this makes all connections appear to come from localhost, unfortunately. Figure \ref{fig:stunnel} depicts the data flow.
meillo@165 522
meillo@172 523 For incoming connections, \NAME{STARTTLS}---defined in \RFC2487---is what \mta{}s implement.
meillo@165 524
meillo@172 525 \masqmail\ is already able to encrypt outgoing connections, but encryption of incoming connections, using \NAME{STARTTLS} should be implemented. This only affects the \SMTP\ server module.
meillo@165 526
meillo@165 527
meillo@165 528
meillo@165 529
meillo@163 530
meillo@163 531 \subsubsection*{Spam prevention}
meillo@163 532
meillo@177 533 ---
meillo@177 534 Spam is a major threat nowadays and the goal is to reduce it to a bearable level (see section \ref{sec:swot-analysis}). Spam fighting is a war are where the good guys tend to lose. Putting too much effort there will result in few gain. Real success will only be possible with new---better---protocols and abandonning the weak legacy technologies. Hence \masqmail\ should be able to provide state-of-the-art spam protection, but not more.
meillo@177 535 ---
meillo@177 536
meillo@173 537 Spam is a major threat to email, as described in section \ref{sec:swot-analysis}. The two main problems are forgable sender addresses and that it is cheap to send hundreds of thousands of messages. Hence, spam senders can operate in disguise and have minimal cost.
meillo@163 538
meillo@173 539 As spam is not just a nuisance for end users, but also for the infrastructure---the \mta{}s---by increasing the amount of mail messages, \MTA{}s need to protect themself. Two approaches are used.
meillo@163 540
meillo@177 541 First refusing spam during the \SMTP\ dialog. This is the way it was meant by the designers of the \SMTP\ protocol. They thought checking the sender and reciptient mail addresses would be enough, but as they are forgable it is not. More and more complex checks need to be done. Checking needs time, but \SMTP\ dialogs time out if it takes too long. Thus only limited time can be used, during the \SMTP\ dialog, for checking if a message seems to be spam. The advantage is that acceptance of bad messages can be simply refused---no responsibility for the message is takes and no further system load is added. See \RFC2505 (especially section 1.5) for detail.
meillo@163 542
meillo@173 543 Second checking for spam after the mail was accepted and queued. Here more processing time can be invested, so more detailed checks can be done. But, as responsibility for messages was taken by accepting them, it is no choice to simply delete spam mail. Checks for spam do not lead to sure results, they just indicate the possibility the message is unwanted mail. \person{Eisentraut} indicates actions to take after a message is recognized as probably spam \cite[pages 18--20]{eisentraut05}. The only acceptable one, for mail the \MTA\ is responsible for, is adding further or rewriting existent header lines. Thus all further work on the message is the same as for non-spam messages.
meillo@165 544
meillo@173 545 Modern \MTA{}s use both techniques in combination. Checks during the \SMTP\ dialog tend to be implemented in the \mta\ to make it fast; checks after the message was queued are often done using external programs (\name{spamassassin} is a well known one). \person{Eisentraut} sees the checks during the \SMTP\ dialog to be essentiell: ``Ganz ohne Analyse während der SMTP-Phase kommt sowieso kein MTA aus, und es ist eine Frage der Einschätzung, wie weit man diese Phase belasten möchte.''\cite[page 25]{eisentraut05} (translated: ``No \MTA\ can go without analysis during the \SMTP\ dialog, anyway, and it is a question of estimation how much to stress this period.'')
meillo@165 546
meillo@173 547 \NAME{DNS} blacklists (short: \NAME{DNSBL}) and \name{greylisting} are checks to be done before accepting the message. Invoking \name{spamassassin}, to add headers containing the estimated spam probability, is best to be invoked after the message is queued.
meillo@165 548
meillo@163 549
meillo@163 550
meillo@163 551
meillo@163 552 \subsubsection*{Virus checking}
meillo@163 553
meillo@173 554 Related to spam is malicous content (short: \name{malware}) like viruses, worms, trojan horses. They, in contrast to spam, do not affect the \MTA\ itself, as they are in the mail body. The same situation in the real world is post offices opening letters to check if they contain something that could harm the recipient. This is not a mail transport concern. Apart of not being the right program to do the job, the \MTA\---the one which is responsible for the recipient---is at a good position to do this work.
meillo@163 555
meillo@177 556 In any way should malware checking be done by external programs that may be invoked by the \mta. But using mail deliver and processing agents, like \name{procmail}, seem to be better suited locations to invoke content scanners.
meillo@163 557
meillo@177 558 A popular email filter framework is \name{amavis} which integrates various spam and virus scanners. The common setup includes a receiving \MTA\ which sends it to \name{amavis} using \SMTP, \name{amavis} processes the mail and sends it then to a second \MTA\ that does the outgoing transfer. \postfix\ and \exim\ can be configured so that one instance can work as both, the \MTA\ for incoming and outgoing transfer. A setup with \sendmail\ needs two separate instances running. It must be quarateed that all mail flows through the scanner.
meillo@163 559
meillo@177 560 A future \masqmail\ would do good to have a single point, where all traffic flows through, that is able to invoke external programs to do mail processing of any kind.
meillo@165 561
meillo@165 562
meillo@177 563 %AMaViS (amavisd-new): email filter framework to integrate spam and virus scanner
meillo@177 564 %\begin{verbatim}
meillo@177 565 %internet -->25 MTA -->10024 amavis -->10025 MTA --> reciptient
meillo@177 566 %| |
meillo@177 567 %+----------------------------+
meillo@177 568 %\end{verbatim}
meillo@177 569 %
meillo@177 570 %postfix and exim can habe both mta servises in the same instance, sendmail needs two instances running.
meillo@177 571 %
meillo@177 572 %MailScanner:
meillo@177 573 %incoming queue --> MailScanner --> outgoing queue
meillo@177 574 %
meillo@177 575 %postfix: with one instance possible, exim and sendmail need two instances running
meillo@165 576
meillo@163 577
meillo@173 578 %message body <-> envelope, header
meillo@173 579 %
meillo@173 580 %anti-virus: clamav
meillo@173 581 %postfix: via amavis
meillo@173 582 %exim: via content-scanning-feature called from acl
meillo@173 583 %sendmail: with milter
meillo@173 584 %procmail
meillo@173 585 %
meillo@173 586 %virus scanner work on file level
meillo@173 587 %amavis receives mail via smtp or pipe, splits it in its parts (MIME) and extracks archives, the come the virus scanners
meillo@173 588 %if the mail is okay, it goes via smtp to a second mta
meillo@173 589
meillo@173 590 %what amavis recognizes:
meillo@173 591 %- invalid headers
meillo@173 592 %- banned files
meillo@173 593 %- viruses
meillo@173 594 %- spam (using spam assassin)
meillo@173 595 %
meillo@173 596 %mimedefang: uses milter interface with sendmail
meillo@173 597
meillo@163 598
meillo@163 599
meillo@163 600 \subsubsection*{Archiving}
meillo@163 601
meillo@177 602 Mail archiving and auditability become more important as electronic mail becomes more important. Ability to archive verbatim copies of every mail coming into and every mail going out of the system, with relation between them, appears to be a goal to achieve.
meillo@163 603
meillo@177 604 \postfix\ for example has a \texttt{always\_bcc} feature, to send a copy of every mail to a definable reciptient. At least this funtionality should be given, although a more complete approach is preferable.
meillo@163 605
meillo@163 606
meillo@163 607
meillo@161 608
meillo@161 609
meillo@185 610 \subsection{The resulting architecture}
meillo@161 611
meillo@178 612 The result is a symetric design, featuring the following parts: Any number of handlers for incoming connections to receive mail and pass it to the module that stores it into the incoming queue. A central scanning module take mail from the incoming queue, processes it in various ways and puts it afterwards into the outgoing queue. Another module takes it out there and passes it to a matching transport module that transfers it to the destination. In other words, three main modules (queue-in, scanning, queue-out) are connected by the two queues (incoming, outgoing); on each end are more modules to receive and send mail---for each protocol one. Figure \ref{fig:masqmail-arch-new} depicts the new designed architecture.
meillo@161 613
meillo@177 614 \begin{figure}
meillo@177 615 \begin{center}
meillo@177 616 \input{input/masqmail-arch-new.tex}
meillo@177 617 \end{center}
meillo@177 618 \caption{A new designed architecture for \masqmail}
meillo@177 619 \label{fig:masqmail-arch-new}
meillo@177 620 \end{figure}
meillo@161 621
meillo@178 622 This architecture is heavily influenced by the ones of \qmail\ and \postfix. Both have different incoming channels that merge in the module that puts mail into the queue; central is the queue (or more of them); and one module takes mail from the queue and passes it to one of the outgoing channels. Mail processing, in any way, is build in in a more explicit way than done in the other two. It is more similar to the \NAME{AR} module of \name{sendmail X}, which is the central point for spam checking.
meillo@178 623
meillo@178 624 Special regard was put on addable support for further mail transfer protocols. This appears to be most similar to \qmail, which was designed to handle multiple protocols.
meillo@178 625 %fixme: do i need all this ``quesses''??
meillo@161 626
meillo@161 627
meillo@185 628 \subsubsection*{Modules and queues}
meillo@161 629
meillo@178 630 The new architecture consists of several modules and two queues. They are defined in more detail now, and the jobs, identified above, are assigned to them. First the three main modules, then the queues, and afterwards the modules for incoming and outgoing transfer.
meillo@161 631
meillo@161 632
meillo@178 633 The \name{queue-in} module creates new spool files in the \name{incoming} queue for incoming messages. It is a process running in background, waiting for connections from one of the receiver modules. When one of them requests for a new spool file, the \name{queue-in} module opens one and returns a positive result. The receiver module then sends the envelope and message, which is written into the spool file by \name{queue-in}. If all went well, another positive result is returend.
meillo@182 634 %fixme: should be no daemon
meillo@178 635
meillo@178 636
meillo@178 637 The \name{scanning} module is the central part of the system. It takes spooled messages from the \name{incoming} queue, works on them, and writes them to the \name{outgoing} queue afterwards (the message is then removed from the \name{incoming} queue, of course). The main job is the processing done on the message. Headers are fixed and missing ones are added if necessary, aliasing is done, and external processing of any kind is triggered. The \name{scanning} module can run in background and look for new mail in regular intvals or signals may be sent to it by \name{queue-in}. Alternatively it can be called by \name{cron}, for example, to do single runs.
meillo@178 638
meillo@178 639
meillo@178 640 The \name{queue-out} module takes messages from the \name{outgoing} queue, queries information about the online connection, and then selects matching routes, creates envelopes for each recipient and passes the messages to the correct transport module. Successfully transfered messages are removed from the \name{outgoing} queue. This module includes some tasks specific to \masqmail.
meillo@178 641
meillo@178 642
meillo@178 643 The \name{incoming} queue stores messages received via one of the incoming channels. The messages are in unprocessed form; only envelope data is prepended.
meillo@178 644
meillo@178 645
meillo@178 646 The \name{outgoing} queue contains processed messages. The header and envelope information is complete and in valid form.
meillo@178 647
meillo@178 648 \name{Receiver modules} are the communication interface between outside senders and the \name{queue-in} module. Each protocol needs a corresponding \name{receiver module} to be supported. Most popular are the \name{sendmail} module (which is a command to be called from the local host) and the \name{smtpd} module (which listens on port 25). Other modules to support other protocols may be added as needed.
meillo@178 649
meillo@178 650 \name{Transport modules}, on the oppersite side of the system, are the modules to send outgoing mail; they are the interface between \name{queue-out} and remote hosts or local commands for further processing. The most popular ones are the \name{smtp} module (which acts as the \SMTP\ client) and the \name{pipe} module (to interface gateways to other systems or networks, like fax or uucp). A module for local delivery is not included, as it is in most other \MTA{}s; the reasons are described in FIXME.%fixme
meillo@178 651 Thus a \name{mail delivery agent} (like \name{procmail}) is to be used with the \name{pipe} module.
meillo@132 652
meillo@132 653
meillo@137 654
meillo@185 655 \subsubsection*{Inter-module communication}
meillo@180 656
meillo@180 657 Communication between modules is required to exchange data and status information. It is also called ``Inter-process communication'' (short: \NAME{IPC}), as modules are programs being part of a larger system, and processes are generally seen as programs in execution.
meillo@180 658
meillo@180 659 The connections between \name{queue-in} and \name{scanning}, aswell as between \name{scanning} and \name{queue-out} is provided by the queues, only sending signals to trigger instant runs may be useful. Communication between receiving and transport modules and the outside world are done using the specific protocol they do handle.
meillo@180 660
meillo@180 661 Left is only communication between the receiver modules and \name{queue-in}, and between \name{queue-out} and the transport modules. Data is exchanged done using \unix\ pipes and a simple protocol is used.
meillo@180 662
meillo@180 663 \begin{figure}
meillo@180 664 \begin{center}
meillo@180 665 \input{input/ipc-protocol.tex}
meillo@180 666 \end{center}
meillo@180 667 \caption{State diagram of the protocol used for \NAME{IPC}}
meillo@180 668 \label{fig:ipc-protocol}
meillo@180 669 \end{figure}
meillo@180 670
meillo@180 671 % timing
meillo@180 672 One dialog consists of the four phases: connection attempt, acceptance reply, data transfer, success reply. The order is always the same. The connection attempt and data transfer are sent by the client process; replies are sent by the server process.
meillo@182 673 %fixme: split between header and data
meillo@180 674
meillo@180 675 % semantics
meillo@180 676 The connection attempt is simply opening the connection. This starts the dialog. A positive reply by the server leads to the data transfer, but a negative reply refuses the connection and resets both client and server to the state before the connection attempt. If the connection attempt was accepted, the client sends the data ending with a terminator sequence. When this terminator appears, the server process knows the complete data was transfered. The server process takes responsibility of the data in sending a positive success reply. A negative success reply resets both client and server to the state before the connection attempt.
meillo@180 677
meillo@180 678 The data transfered needs to be of specific format. Used is the same format in which messages are spooled in the mail queues. See the following section for details. %fixme: check if it is the following section
meillo@182 679 %fixme: split between header and data
meillo@180 680
meillo@180 681 % syntax
meillo@180 682 Data transfer is done sending plain text data. %fixme: utf8 ?
meillo@182 683 The terminator sequence used to indicate the end of the data transfer is a single dot on a line on its own. Line separators are the combination of \name{Carriage Return} and \name{Line Feed}, as it is used in various Internet protocols like \SMTP. Replys are one-digit numbers with \texttt{0} meaning success and any other number (\texttt{1}--\texttt{9}) indicate failure. %fixme: What are the octal values?
meillo@182 684 %fixme: split between header and data
meillo@180 685
meillo@180 686 Figure \ref{fig:ipc-protocol} is a state diagram for the protocol.
meillo@137 687
meillo@149 688
meillo@149 689
meillo@185 690 \subsubsection*{Spool file format}
meillo@177 691
meillo@182 692 The spool file format is basically the same as the one in current \masqmail: one file for the message body, the other for envelope and header information. The data file is stored in a separate data pool. It is written by \name{queue-in}, \name{scanning} can read it if necessary, \name{queue-out} reads it to generate the outgoing message, and deletes it after successful transfer. The header file (including the envelope) is written into the \name{incoming} queue. The \name{scanning} modules reads it, processes it, and writes a modified copy into the \name{outgoing} queue; the file in \name{incoming} is deleted then. \name{queue-out} finally takes the header file from \name{outgoing} to generate the resulting message. This data flow is shown in figure \ref{fig:queue-data-flow}.
meillo@180 693
meillo@182 694 \begin{figure}
meillo@182 695 \begin{center}
meillo@182 696 \input{input/queue-data-flow.tex}
meillo@182 697 \end{center}
meillo@182 698 \caption{Data flow of messages in the queue}
meillo@182 699 \label{fig:queue-data-flow}
meillo@182 700 \end{figure}
meillo@180 701
meillo@182 702 The queue consists of three directories within the queue path. Two, named \name{incoming} and \name{outgoing}, for storing the header files; one, called \name{pool}, to store the message bodies. The files being part of one message share the same unique name. The header files internal structure can be the same as the one of current \masqmail.
meillo@180 703
meillo@182 704 Messages in queues are a header file in \name{incoming} or \name{outgoing} and a data file in \name{pool}. The header file owner's executable bit indicates if the file is ready for further processing: the module that writes the file into the queue sets the bit as last action. Modules that read from the queue can process messages with the bit set.
meillo@182 705
meillo@182 706 No spool files are modified after they are written to disk. Modifications to header files can be made by the \name{scanning} module in the ``move'' from \name{incoming} to \name{outgoing}---it is a create and remove, actually. Further rewriting can happen in \name{queue-out}, as well without altering the file.
meillo@182 707
meillo@182 708 Data files do not change at all within the system. They are written in default local plain text format. Required translation is done in the receiver and transport modules.
meillo@182 709
meillo@182 710
meillo@185 711 \begin{tabular}[hbt]{ l l }
meillo@185 712
meillo@185 713 \mbox{ queue-in:} & \mbox{
meillo@185 714 \begin{tabular}[hbt]{| c | c | c |}
meillo@185 715 \hline
meillo@185 716 incoming & outgoing & pool \\
meillo@185 717 \hline
meillo@185 718 \hline
meillo@185 719 - & - & - \\
meillo@185 720 \hline
meillo@185 721 0600 & - & - \\
meillo@185 722 \hline
meillo@185 723 0600 & - & 0600 \\
meillo@185 724 \hline
meillo@185 725 0700 & - & 0600 \\
meillo@185 726 \hline
meillo@185 727 \end{tabular}
meillo@185 728 } \\
meillo@185 729
meillo@185 730 \quad & \\
meillo@185 731
meillo@185 732 \mbox{scanning:} & \mbox{
meillo@185 733 \begin{tabular}[hbt]{| c | c | c |}
meillo@185 734 \hline
meillo@185 735 incoming & outgoing & pool \\
meillo@185 736 \hline
meillo@185 737 \hline
meillo@185 738 0700 & - & 0600 \\
meillo@185 739 \hline
meillo@185 740 0700 & 0600 & 0600 \\
meillo@185 741 \hline
meillo@185 742 0700 & 0700 & 0600 \\
meillo@185 743 \hline
meillo@185 744 - & 0700 & 0600 \\
meillo@185 745 \hline
meillo@185 746 \end{tabular}
meillo@185 747 } \\
meillo@185 748
meillo@185 749 \quad & \\
meillo@185 750
meillo@185 751 \mbox{queue-out:} & \mbox{
meillo@185 752 \begin{tabular}[hbt]{| c | c | c |}
meillo@185 753 \hline
meillo@185 754 incoming & outgoing & pool \\
meillo@185 755 \hline
meillo@185 756 \hline
meillo@185 757 - & 0700 & 0600 \\
meillo@185 758 \hline
meillo@185 759 - & 0700 & - \\
meillo@185 760 \hline
meillo@185 761 - & - & - \\
meillo@185 762 \hline
meillo@185 763 \end{tabular}
meillo@185 764 } \\
meillo@185 765
meillo@185 766 \end{tabular}
meillo@182 767
meillo@182 768 A sample header file.
meillo@180 769 \begin{verbatim}
meillo@182 770 1LGtYh-0ut-00 (backup copy of the file name)
meillo@182 771 MF:<meillo@dream> (envelope: sender)
meillo@182 772 RT: <user@example.org> (envelope: recipient)
meillo@182 773 PR:local (meta info: protocol)
meillo@182 774 ID:meillo (meta info: id/user/ip)
meillo@182 775 DS: 18 (meta info: size)
meillo@182 776 TR: 1230462707 (meta info: timestamp)
meillo@182 777 (following: headers)
meillo@182 778 HD:Received: from meillo by dream with local (masqmail 0.2.21) id
meillo@182 779 1LGtYh-0ut-00 for <user@example.org>; Sun, 28 Dec 2008 12:11:47 +0100
meillo@182 780 HD:To: user@example.org
meillo@182 781 HD:Subject: test mail
meillo@182 782 HD:From: <meillo@dream>
meillo@182 783 HD:Date: Sun, 28 Dec 2008 12:11:47 +0100
meillo@182 784 HD:Message-ID: <1LGtYh-0ut-00@dream>
meillo@180 785 \end{verbatim}
meillo@180 786
meillo@180 787
meillo@177 788
meillo@182 789
meillo@185 790 \subsubsection*{Rights and permission}
meillo@177 791
meillo@182 792 The user set required for \qmail\ seems to be too complex. One special user, like \postfix\ uses, is more appropriate. \name{root} privilege and \name{setuid} permission is avoided as much as possible.
meillo@182 793
meillo@182 794 Table \ref{tab:new-masqmail-permissions} shows the suggested ownership and permissions of the modules. Figure \ref{fig:new-masqmail-queue} shows the permissions and ownership used for the queue.
meillo@182 795
meillo@182 796 \begin{table}
meillo@182 797 \begin{center}
meillo@182 798 \input{input/new-masqmail-permissions.tex}
meillo@182 799 \end{center}
meillo@182 800 \caption{Ownership and permissions of the modules}
meillo@182 801 \label{tab:new-masqmail-permission}
meillo@182 802 \end{table}
meillo@182 803
meillo@182 804 \begin{figure}
meillo@182 805 \begin{center}
meillo@182 806 \input{input/new-masqmail-queue.tex}
meillo@182 807 \end{center}
meillo@182 808 \caption{Ownership and permissions of the queue}
meillo@182 809 \label{fig:new-masqmail-queue}
meillo@182 810 \end{figure}
meillo@182 811
meillo@182 812
meillo@182 813
meillo@182 814
meillo@182 815
meillo@180 816 setuid/setgid or not?
meillo@149 817
meillo@180 818 what can crash if an attacker succeeds?
meillo@180 819
meillo@180 820 where to drop privelege?
meillo@180 821
meillo@180 822 how is which process invoked?
meillo@180 823
meillo@180 824 master process? needed, or wanted?
meillo@180 825
meillo@180 826 which are the daemon processes?
meillo@149 827
meillo@149 828
meillo@149 829
meillo@149 830
meillo@149 831
meillo@93 832
meillo@93 833
meillo@185 834 http://fanf.livejournal.com/50917.html %how not to design an mta - the sendmail command
meillo@185 835 http://fanf.livejournal.com/51349.html %how not to design an mta - partitioning for security
meillo@185 836 http://fanf.livejournal.com/61132.html %how not to design an mta - local delivery
meillo@185 837 http://fanf.livejournal.com/64941.html %how not to design an mta - spool file format
meillo@185 838 http://fanf.livejournal.com/65203.html %how not to design an mta - spool file logistics
meillo@185 839 http://fanf.livejournal.com/65911.html %how not to design an mta - more about log-structured MTA queues
meillo@185 840 http://fanf.livejournal.com/67297.html %how not to design an mta - more log-structured MTA queues
meillo@185 841 http://fanf.livejournal.com/70432.html %how not to design an mta - address verification
meillo@185 842 http://fanf.livejournal.com/72258.html %how not to design an mta - content scanning
meillo@185 843
meillo@185 844
meillo@185 845