1.1 --- /dev/null	Thu Jan 01 00:00:00 1970 +0000
     1.2 +++ b/docs/imaprc.txt	Mon Sep 14 15:17:45 2009 +0900
     1.3 @@ -0,0 +1,613 @@
     1.4 +/* ========================================================================
     1.5 + * Copyright 1988-2006 University of Washington
     1.6 + *
     1.7 + * Licensed under the Apache License, Version 2.0 (the "License");
     1.8 + * you may not use this file except in compliance with the License.
     1.9 + * You may obtain a copy of the License at
    1.10 + *
    1.11 + *     http://www.apache.org/licenses/LICENSE-2.0
    1.12 + *
    1.13 + * 
    1.14 + * ========================================================================
    1.15 + */
    1.16 +
    1.17 +		       .imaprc secrets revealed!
    1.18 +		      Mark Crispin, June 17, 2002
    1.19 +
    1.20 +The following information describes the format of the /etc/c-client.cf
    1.21 +and ~/.imaprc file.  The Columbia MM ~/.mminit file is also read by
    1.22 +c-client; however, the only command that ~/.mminit has in common is
    1.23 +set keywords.
    1.24 +
    1.25 +**********************************************************************
    1.26 +*		     DANGER!  BEWARE!  TAKE CARE!		     *
    1.27 +**********************************************************************
    1.28 +*								     *
    1.29 +*  These files, and this documentation, are for internal UW usage    *
    1.30 +* only.  This capability is for UW experimental tinkering, and most  *
    1.31 +* emphatically *not* for sorcerer's apprentices at other sites who   *
    1.32 +* feel that if a config file capability exists, they must write a    *
    1.33 +* config file whether or not there is any need for one.		     *
    1.34 +*								     *
    1.35 +*  This information is subject to change without notice.  Commands   *
    1.36 +* may be added, removed, or altered.  The behavior of comamnds may   *
    1.37 +* change.  Do not use any of this information without consulting me  *
    1.38 +* first.  c-client's defaults have been carefully chosen to be right *
    1.39 +* for general-purpose and most special-purpose configurations.  If   *
    1.40 +* you tinker with these defaults, all hell may break loose.	     *
    1.41 +*								     *
    1.42 +*  This is not an idle threat.  There have been several instances of *
    1.43 +* people who ignored these warnings and have gotten burned.	     *
    1.44 +*								     *
    1.45 +*  Don't even trust this file to work.  Many of the things which can *
    1.46 +* be changed by this file can also be changed by the application,    *
    1.47 +* and it is totally unpredictable which will take precedence.  It    *
    1.48 +* all depends upon how the application is coded.  Not only that, you *
    1.49 +* may cause the application to crash.                                *
    1.50 +*								     *
    1.51 +*  In other words, keep your cotton-pickin' hands off my defaults.   *
    1.52 +* If it crashes and erases your mail, I don't want to hear about it. *
    1.53 +* Consider 'em ``mandatory defaults''.  Got a nice ring, eh?  :-) If *
    1.54 +* you must tinker with defaults, play with the .pinerc and pine.conf *
    1.55 +* files in Pine.  It's got options galore, all supported for you to  *
    1.56 +* have fun.  They're also documented; so well documented, it takes   *
    1.57 +* two strong men to carry around all the documentation.	 ;-) ;-)     *
    1.58 +*								     *
    1.59 +*  Joking aside, you really shouldn't be fooling around with this    *
    1.60 +* capability.  It's dangerous, and you can shoot yourself in the     *
    1.61 +* foot easily.  If you need custom changes, you are better off with  *
    1.62 +* local source code modifications.  Seriously.			     *
    1.63 +*								     *
    1.64 +*  One last warning: don't believe anything that you read in this    *
    1.65 +* document.  Every effort has been made to ensure that this document *
    1.66 +* is incomplete and inaccurate, and I take no responsibility for any *
    1.67 +* glimmers of correct information that may, by some fluke, be here.  *
    1.68 +*								     *
    1.69 +**********************************************************************
    1.70 +
    1.71 +The files are read in order: /etc/c-client.cf, ~/.mminit, ~/.imaprc,
    1.72 +and an entry in a later file overrides the setting of an earlier file
    1.73 +except as noted below.  This ordering and overriding behavior may
    1.74 +change without notice.
    1.75 +
    1.76 +Almost all of these facilities can also be set via the mail_parameters()
    1.77 +call in the program.  Whether the file overrides mail_parameters(), or
    1.78 +mail_parameters() overrides the file, is indeterminate.  It will vary
    1.79 +from program to program, and it may be one way in one version and the
    1.80 +other way in the next version.  It's completely unpredictable, and so
    1.81 +anything you do with these files has to be in complete knowledge of what
    1.82 +the version of each program you're running is going to do.  This is
    1.83 +because the files do something for testing, but the real capability for
    1.84 +configurability is put in the program instead.  Are you getting the
    1.85 +feeling that you shouldn't be messing with these files yet?
    1.86 +
    1.87 +The very first line of the file MUST start with the exact string "I
    1.88 +accept the risk".  This ensures that you have checked the file for
    1.89 +correctness against this version of the IMAP toolkit.  This enable
    1.90 +string may change without notice in future versions, and the new
    1.91 +string may or may not be accurately described in an updated version of
    1.92 +this file.  So any time you install software that uses the IMAP
    1.93 +toolkit, you need to check the new version against these files (if you
    1.94 +have insisted upon creating them in spite of all warnings).  If two
    1.95 +pieces of software use different versions of the IMAP toolkit with
    1.96 +incompatible requirements, one of them won't work.  Re-read the
    1.97 +warning above about why you should not use these files.
    1.98 +
    1.99 +Subsequent lines are read from the file one at a time.  Case does not
   1.100 +matter.  Unrecognized commands are ignored.
   1.101 +
   1.102 +1) set new-folder-format
   1.103 +   sets what format new mailboxes are created in.  This also controls
   1.104 +   default delivery via tmail and dmail.
   1.105 +
   1.106 +   a) set new-folder-format same-as-inbox
   1.107 +      Folder is created using the same mailbox format as INBOX.  If
   1.108 +      INBOX is empty, it defaults to system standard.
   1.109 +
   1.110 +   b) set new-folder-format system-standard
   1.111 +      This is the default.  Folder is created using the wired-in system
   1.112 +      standard format, which on most UNIX systems is ordinary UNIX
   1.113 +      /bin/mail format.  On SCO systems, this is MMDF.
   1.114 +
   1.115 +   c) set new-folder-format <driver name>
   1.116 +      Folder is created using the given driver name, e.g. mbx, unix,
   1.117 +      mmdf, etc.
   1.118 +
   1.119 +   There is no protection against setting this to a silly value (e.g.
   1.120 +   news, nntp, dummy) and doing so is a great way to screw things up.
   1.121 +   Setting this to mh does not do what you think it does.  Setting this
   1.122 +   to tenex or mtx isn't particularly useful.
   1.123 +
   1.124 +2) set empty-folder-format
   1.125 +   sets what format data is written into an empty mailbox file using
   1.126 +   mail_copy() or mail_append().  This also controls default delivery
   1.127 +   via tmail.
   1.128 +
   1.129 +   a) set empty-folder-format same-as-inbox
   1.130 +      Data is written using the same mailbox format as INBOX.  If
   1.131 +      INBOX is empty, it defaults to system standard.
   1.132 +
   1.133 +   b) set empty-folder-format system-standard
   1.134 +      This is the default.  Data is written using the wired-in system
   1.135 +      standard format, which on most UNIX systems is ordinary UNIX
   1.136 +      /bin/mail format.  On SCO systems, this is MMDF.
   1.137 +
   1.138 +   c) set-empty-folder-format <driver name>
   1.139 +      Data is written using the given driver name, e.g. tenex, unix,
   1.140 +      mmdf, etc.
   1.141 +
   1.142 +   There is no protection against setting this to a silly value (e.g.
   1.143 +   news, nntp, dummy) and doing so is a great way to screw things up.
   1.144 +   Setting this to mh, mbx, or mx does not work.
   1.145 +
   1.146 +3) set keywords <word1>, <word2>, ... <wordn>
   1.147 +   Sets the list of keyword flags (supported by tenex and mtx) to the
   1.148 +   given list.  Up to 30 flags may be given.  Since these names
   1.149 +   correspond to numeric bits, the order of the keywords can not be
   1.150 +   changed, nor can keywords be removed or inserted (you can append
   1.151 +   new keywords, up to the limit of 30).
   1.152 +
   1.153 +   Set keywords is a deprecated command.  It may not appear in
   1.154 +   future versions, or it may appear in a changed form.  It exists
   1.155 +   only for compatibility with MM, and should only appear in ~/.mminit
   1.156 +   and not in the other files.  It is likely to disappear entirely in
   1.157 +   IMAP4.
   1.158 +
   1.159 +   There is no protection against setting these to silly values, and
   1.160 +   doing so is a great way to cause a crash.
   1.161 +
   1.162 +4) set from-widget header-only
   1.163 +   Sets smart insertion of the > character in front of lines that
   1.164 +   begin with ``From ''.  Only such lines that are also in UNIX mbox
   1.165 +   header file format will have a > character inserted.  The default
   1.166 +   is to insert the > character in front of all lines which begin with
   1.167 +   ``From '', for the benefit of legacy tools that get confused
   1.168 +   otherwise.
   1.169 +
   1.170 +5) set black-box-directory <directory name>
   1.171 +   Sets the directory in which the user's data can be found.  A user's
   1.172 +   folders can be found in a subdirectory of the black box directory
   1.173 +   named with the user's username.  For example, if the blackbox
   1.174 +   directory is /usr/spool/folders/, user jones' data can be found
   1.175 +   in /usr/spool/folders/jones/.  The user's black-box directory is
   1.176 +   the location of folders, .mminit, .imaprc, .newsrc, and all other
   1.177 +   files used by c-client; internally, it sets c-client's idea of the
   1.178 +   user's ``home directory'', overriding /etc/passwd.
   1.179 +
   1.180 +   This command may not appear in ~/.mminit or ~/.imaprc
   1.181 +
   1.182 +   In black-box mode, it is not permitted to access any folders
   1.183 +   outside of the user's personal blackbox directory.  The breakouts
   1.184 +   ``/'', ``~'', and ``..'' are not permitted.
   1.185 +
   1.186 +   In order to make this work without crashing, you must set another
   1.187 +   option which is not listed in this document.
   1.188 +
   1.189 +   There is no protection against setting this to a silly value, and
   1.190 +   doing so is a great way to cause a crash.
   1.191 +
   1.192 +6) set local-host <host name>
   1.193 +   Sets c-client's idea of the local host name.
   1.194 +
   1.195 +   There is no protection against setting this to a silly value, and
   1.196 +   doing so is a great way to cause a crash.
   1.197 +
   1.198 +7) set news-active-file <file name>
   1.199 +   Sets the location of the news active file, if it is not in the
   1.200 +   standard place.
   1.201 +
   1.202 +   It is recommended to use a courtesy symbolic link instead.
   1.203 +
   1.204 +   There is no protection against setting this to a silly value, and
   1.205 +   doing so is a great way to cause a crash.
   1.206 +
   1.207 +8) set news-spool-directory <directory name>
   1.208 +   Sets the location of the news spool, if it is not in the standard
   1.209 +   place.
   1.210 +
   1.211 +   It is recommended to use a courtesy symbolic link instead.
   1.212 +
   1.213 +   There is no protection against setting this to a silly value, and
   1.214 +   doing so is a great way to cause a crash.
   1.215 +
   1.216 +9) set news-state-file <file name>
   1.217 +   Sets the location of the news state file (normally $(USER)/.newsrc).
   1.218 +
   1.219 +   This is not very useful in /etc/c-client.cf because it is a file name.
   1.220 +   Setting this in /etc/c-client.cf would set all users to the same file
   1.221 +   as their newsrc, which is probably not what you want.
   1.222 +
   1.223 +   There is no protection against setting this to a silly value, and
   1.224 +   doing so is a great way to cause a crash.
   1.225 +
   1.226 +10) set system-inbox <file name>
   1.227 +   Sets the location of the "system inbox", if it is not in the standard
   1.228 +   place.  This is the default location of INBOX, or the mail drop point
   1.229 +   from which mail is snarfed (e.g. in tenex, mtx, mbox, mh formats).
   1.230 +
   1.231 +   This is not very useful in /etc/c-client.cf because it is a file name.
   1.232 +   Setting this in /etc/c-client.cf would set all users to the same file
   1.233 +   as their system inbox, which is probably not what you want.
   1.234 +
   1.235 +   There is no protection against setting this to a silly value, and
   1.236 +   doing so is a great way to cause a crash.
   1.237 +
   1.238 +11) set tcp-open-timeout <number>
   1.239 +    Sets the number of seconds that the TCP routines will block on opening
   1.240 +    a TCP connection before timing out.  If a timeout occurs, the connection
   1.241 +    attempt is aborted.
   1.242 +
   1.243 +    The default is zero, meaning use the operating system default (75
   1.244 +    seconds on most UNIX systems).
   1.245 +
   1.246 +    There is no protection against setting this to an excessively small
   1.247 +    value, such as 1, and doing so is a great way to cause users extreme
   1.248 +    grief.
   1.249 +
   1.250 +12) set tcp-read-timeout <number>
   1.251 +    Sets the number of seconds that the TCP routines will block on reading
   1.252 +    data before calling the timeout routine.  If no timeout routine is set
   1.253 +    by the program, the connection will be aborted on a timeout.
   1.254 +
   1.255 +    The default is zero, meaning infinite.
   1.256 +
   1.257 +    There is no protection against setting this to an excessively small
   1.258 +    value, such as 1, and doing so is a great way to cause users extreme
   1.259 +    grief.
   1.260 +
   1.261 +13) set tcp-write-timeout <number>
   1.262 +    Sets the number of seconds that the TCP routines will block on sending
   1.263 +    data before calling the timeout routine.  If no timeout routine is set
   1.264 +    by the program, the connection will be aborted on a timeout.
   1.265 +
   1.266 +    The default is zero, meaning infinite.
   1.267 +
   1.268 +    There is no protection against setting this to an excessively small
   1.269 +    value, such as 1, and doing so is a great way to cause users extreme
   1.270 +    grief.
   1.271 +
   1.272 +14) set rsh-timeout <number>
   1.273 +    Sets the number of seconds that the rsh routines will block on opening
   1.274 +    an rimapd connection before timing out.  If a timeout occurs, the
   1.275 +    rsh connection attempt is aborted.  A zero timeout will disable rsh.
   1.276 +
   1.277 +    The default is 15 seconds.
   1.278 +
   1.279 +    There is no protection against setting this to an excessively small
   1.280 +    value, such as 1, and doing so is a great way to cause users extreme
   1.281 +    grief.
   1.282 +
   1.283 +15) set maximum-login-trials <number>
   1.284 +    Sets the number of iterations of asking the user, via mm_login(), for
   1.285 +    a user name and password, before cancelling the attempt.
   1.286 +
   1.287 +    The default is 3.
   1.288 +
   1.289 +    There is no protection against setting this to zero, and doing so is
   1.290 +    a great way to cause users extreme grief.
   1.291 +
   1.292 +16) set lookahead <number>
   1.293 +    Sets the number of envelopes that are looked ahead in IMAP, in
   1.294 +    mail_fetchstructure().  This is based on the guess that in such
   1.295 +    operations as drawing browser lines, if you get data for message n
   1.296 +    you are likely to want it for message n+1, n+2,... in short order.
   1.297 +    Lookahead preloads the c-client  cache and saves unnecessary RTTs.
   1.298 +
   1.299 +    The default is 20, a good number for a browser on a 24x80 screen, and
   1.300 +    small enough to usually have no significant real-time difference from
   1.301 +    a single message fetch.
   1.302 +
   1.303 +    Setting it to 0 turns off lookahead.
   1.304 +
   1.305 +    There is no protection against setting this ridiculously high and
   1.306 +    incurring performance penalties as a result.
   1.307 +
   1.308 +17) set prefetch <number>
   1.309 +    Sets the number of envelops which are automatically fetched for the
   1.310 +    messages which match in a search.  This is based on the guess that
   1.311 +    in a browser that is "zoomed" on the results of a search, you are
   1.312 +    likely to want the envelope data for each of those messages in
   1.313 +    short order.  Prefetching reloads the c-client cache, saves
   1.314 +    unnecessary RTTs, and avoids loading undesired envelopes due to
   1.315 +    lookahead (see above).
   1.316 +
   1.317 +    The default is 20.
   1.318 +
   1.319 +    Setting it to 0 turns off prefetch.
   1.320 +
   1.321 +    There is no protection against setting this ridiculously high and
   1.322 +    incurring performance penalties as a result.
   1.323 +
   1.324 +18) set close-on-error <number>
   1.325 +    If non-zero, IMAP connections are closed if an EXAMINE or SELECT
   1.326 +    command fails.  Otherwise, they are left half-open, and can be used
   1.327 +    again to select some other mailbox.  The mailbox name in the stream
   1.328 +    is set to {serverhost}<no_mailbox>
   1.329 +
   1.330 +    The default is zero (do not close on error).
   1.331 +
   1.332 +19) set imap-port <number>
   1.333 +    Set the TCP/IP contact port to use for IMAP.  This overrides the
   1.334 +    wired-in setting and the setting from /etc/services, and can in
   1.335 +    turn be overridden by an explicit user specification in the mailbox
   1.336 +    name, e.g. {serverhost:143}foo
   1.337 +
   1.338 +    The default is zero (use setting from /etc/services or the wired-in
   1.339 +    setting (143).
   1.340 +
   1.341 +    There is no protection against setting this to a silly value, and
   1.342 +    doing so is a great way to cause users extreme grief.
   1.343 +
   1.344 +20) set pop3-port <number>
   1.345 +    Set the TCP/IP contact port to use for POP3.  This overrides the
   1.346 +    wired-in setting and the setting from /etc/services, and can in
   1.347 +    turn be overridden by an explicit user specification in the mailbox
   1.348 +    name, e.g. {serverhost:110/pop3}
   1.349 +
   1.350 +    The default is zero (use setting from /etc/services or the wired-in
   1.351 +    setting (110).
   1.352 +
   1.353 +    There is no protection against setting this to a silly value, and
   1.354 +    doing so is a great way to cause users extreme grief.
   1.355 +
   1.356 +21) set uid-lookahead <number>
   1.357 +    Sets the number of UIDs that are looked ahead in IMAP in mail_uid().
   1.358 +    Lookahead preloads the c-client cache and saves unnecessary RTTs.
   1.359 +
   1.360 +    The default is 1000, small enough to usually have no significant
   1.361 +    real-time difference from a single message UID fetch.
   1.362 +
   1.363 +    Setting it to 0 turns off lookahead.
   1.364 +
   1.365 +    There is no protection against setting this ridiculously high and
   1.366 +    incurring performance penalties as a result.
   1.367 +
   1.368 +22) set mailbox-protection <number>
   1.369 +    Set the default protection for newly-created mailbox files.
   1.370 +
   1.371 +    The default is 384.
   1.372 +
   1.373 +    There is no protection against setting this to a silly value, and
   1.374 +    doing so is a great way to screw things up massively.
   1.375 +
   1.376 +23) set directory-protection <number>
   1.377 +    Set the default protection for newly-created directories.
   1.378 +
   1.379 +    The default is 448.
   1.380 +
   1.381 +    There is no protection against setting this to a silly value, and
   1.382 +    doing so is a great way to screw things up massively.
   1.383 +
   1.384 +24) set lock-protection <number>
   1.385 +    Set the default protection for lock files
   1.386 +
   1.387 +    The default is 438, which is necessary if locks are to be respected
   1.388 +    by processes running as other UIDs.
   1.389 +
   1.390 +    There is no protection against setting this to a silly value, and
   1.391 +    contrary to what you may think just about any value other than 438
   1.392 +    turns out to be a silly value.
   1.393 +
   1.394 +25) set disable-fcntl-locking <number>
   1.395 +    This only applies to SVR4 systems.
   1.396 +
   1.397 +    If non-zero, fnctl() locking is not attempted.  In the past, this
   1.398 +    was used to avoid locking NFS files.  If NFS is involved, the evil
   1.399 +    lockd/statd daemons get invoked.  These daemons supposedly work over
   1.400 +    NFS, but really don't.
   1.401 +
   1.402 +    You probably don't really want to do this, though, because now the
   1.403 +    flock() emulator (which calls fcntl()) now checks to see if the file
   1.404 +    is accessed via NFS and no-ops the lock.  This is compatible with
   1.405 +    BSD.
   1.406 +
   1.407 +    Disabling fcntl() locking loses a great deal of locking protection
   1.408 +    on local files as well as NFS files (which now never have locking
   1.409 +    protection).
   1.410 +
   1.411 +    The default is zero (fcntl() locking is enabled).
   1.412 +
   1.413 +26) set lock-EACCES-error <number>
   1.414 +    If non-zero, a warning message is given if an attempt to create a
   1.415 +    lock file fails.  Otherwise, EACCES is treated as a "silent failure",
   1.416 +    and it proceeds without trying to use the lock file.  This is for
   1.417 +    the benefit of users on systems with paranoid /usr/spool/mail
   1.418 +    protections which don't let users create /usr/spool/mail/$(USER).lock
   1.419 +    files; these unfortunate users would be harassed with a flood of
   1.420 +    error messages otherwise.  The problem is that on SVR4, if EACCES
   1.421 +    remains disabled and fcntl() locking is also disabled, then there is
   1.422 +    no locking at all which is doubleplus-ungood.
   1.423 +
   1.424 +    If the site is paranoid on /usr/spool/mail protections AND if there
   1.425 +    is no fcntl() locking (SVR4) or usable flock() locking (e.g. NFS),
   1.426 +    then there is no way to win.  Find a different system to use.
   1.427 +
   1.428 +    The default is non-zero (report EACCESS as an error).
   1.429 +
   1.430 +27) set list-maximum-level <number>
   1.431 +    Sets the maximum depth of recursion that a * wildcard list will go
   1.432 +    down the directory tree.  0 means that no recursion is permitted,
   1.433 +    and * becomes like %.
   1.434 +
   1.435 +    The default is 20.
   1.436 +
   1.437 +    There is no protection against setting this to a ridiculously high
   1.438 +    value.  Since LIST will follow symbolic links, it can effectively
   1.439 +    recurse infinitely, until the name strings get large enough that
   1.440 +    some name limit is exceeded.
   1.441 +
   1.442 +28) set anonymous-home-directory <directory name>
   1.443 +   Sets the location of the anonymous home directory, if it is not in
   1.444 +   the standard  place.
   1.445 +
   1.446 +   It is recommended to use a courtesy symbolic link instead.
   1.447 +
   1.448 +   There is no protection against setting this to a silly value, and
   1.449 +   doing so is a great way to cause a crash.
   1.450 +
   1.451 +29) set chroot-server <number>
   1.452 +   This option is for closed server systems only.  If defined, a chroot()
   1.453 +   call to the user's home directory is done as part of the login
   1.454 +   process.  This has the effect of preventing access to any files
   1.455 +   outside of the user's home directory (including shared mailboxes).
   1.456 +
   1.457 +   Shared mailboxes with other users can't possibly work with this
   1.458 +   option, because there is no way to export lock information to other
   1.459 +   users.
   1.460 +
   1.461 +   This should be done ONLY on systems which do not permit users to
   1.462 +   have shell access
   1.463 +
   1.464 +   This option should NEVER(!!) be set if users are allowed shell access.
   1.465 +   Doing so actually makes the system *less* secure, since the user could
   1.466 +   create an etc subdirectory which would be treated as real /etc by such
   1.467 +   programs as /bin/su.
   1.468 +
   1.469 +   The default is zero (don't do chroot).
   1.470 +
   1.471 +   This option is strongly *NOT* recommended.
   1.472 +
   1.473 +30) set disable-automatic-shared-namespaces <number>
   1.474 +   Never look up the "ftp", "imappublic", and "imapshared" users as
   1.475 +   posssible home directories for the #ftp, #public, and #shared
   1.476 +   namespaces.  On some systems (reportedly including AIX 4.3.3)
   1.477 +   getpwnam() of an unknown user name is horrendously slow.
   1.478 +
   1.479 +   Note that this does not remove the #ftp, #public, and #shared
   1.480 +   namespaces, and they can still be set up by other means.
   1.481 +
   1.482 +   The default is zero (shared namespaces are automatic).
   1.483 +
   1.484 +31) set advertise-the-world <number>
   1.485 +   Include the UNIX root as a shared namespace.  This is generally a bad
   1.486 +   idea, since certain IMAP clients (names withheld to protect the guilty)
   1.487 +   will take this as license to download the entire filesystem tree.
   1.488 +
   1.489 +   The default is zero (don't advertise the world).
   1.490 +
   1.491 +32) set mail-subdirectory <subdirectory name>
   1.492 +   Change the default connected directory from the user's home directory
   1.493 +   to the named subdirectory of the user's home directory.  For example,
   1.494 +   setting MAILSUBDIR="mail" will cause the POP2 and IMAP servers to
   1.495 +   connect to the user's ~/mail subdirectory.  This is equivalent to
   1.496 +   the env_unix.c edit described in Example 2 of the CONFIG file.
   1.497 +
   1.498 +   Note that if the subdirectory does not exist, the result is undefined.
   1.499 +   It is probably an extremely bad idea to set this unless you can
   1.500 +   guarantee that the subdirectory exists for all users.  If you can not
   1.501 +   guarantee this, then you should leave the default as the user's home
   1.502 +   directory and allow them to configure a personal default in their IMAP
   1.503 +   client.
   1.504 +
   1.505 +   The default is not to use any subdirectory.
   1.506 +
   1.507 +33) set allow-user-config <number>
   1.508 +   Allow users to use ~/.imaprc and ~/.mminit files.
   1.509 +
   1.510 +   The default is zero (don't allow user config files).
   1.511 +
   1.512 +34) set allow-reverse-dns <number>
   1.513 +   By default, the servers (ipop[23]d and imapd) will do gethostbyaddr()
   1.514 +   on the local and remote sockets so that imapd can identify itself
   1.515 +   properly (this is important when the same CPU hosts multiple virtual
   1.516 +   hosts on different IP addresss) and also includes the client's name
   1.517 +   when it writes to the syslog.  There are also client gethostbyaddr()
   1.518 +   calls, used primarily by authentication mechanisms.
   1.519 +
   1.520 +   Setting this option to zero disables all gethostbyaddr() calls.  The
   1.521 +   returned "host name" string for the socket is just the bracketed
   1.522 +   [12.34.56.78] form, as if the reverse DNS lookup failed.
   1.523 +
   1.524 +   WARNING: Some authentication mechanisms, e.g. Kerberos V, depend upon
   1.525 +   the host names being right, and if you set this option, it won't work.
   1.526 +
   1.527 +   You should only do this if you are encountering server performance
   1.528 +   problems due to a misconfigured DNS, e.g. long startup delays or
   1.529 +   client timeouts.
   1.530 +
   1.531 +   The default is non-zero (allow reverse DNS).
   1.532 +
   1.533 +35) set disable-plaintext <number>
   1.534 +   Disable plaintext password authentication (LOGIN command, AUTH=LOGIN,
   1.535 +   and AUTH=PLAIN).
   1.536 +
   1.537 +   The default is zero (allow plaintext authentication).
   1.538 +
   1.539 +36) set trust-dns <number>
   1.540 +   By default, host names are canonicalized via gethostbyname() for
   1.541 +    everything except for SSL certificate validation.
   1.542 +
   1.543 +   This can represent a security bug due to DNS spoofing, but is more
   1.544 +    likely to deliver results that users expect.  It also may be necessary
   1.545 +    for SASL authentication to work right (e.g. generating a correct name
   1.546 +    for a Kerberos service principal) if the name entered by the user is a
   1.547 +    CNAME or not a fully-qualified domain name.
   1.548 +
   1.549 +   If trust-dns is set to zero, no host name canonicalization is done.
   1.550 +    The user's actual entered name is used for SASL authentication and
   1.551 +    will appear in the mailbox name of the open stream.
   1.552 +
   1.553 +   The default is non-zero (do DNS canonicalization).
   1.554 +
   1.555 +37) set sasl-uses-ptr-name <number>
   1.556 +   By default, if trust-dns is set, the host names used in authentication
   1.557 +    (e.g. to generate a Kerberos service principal) are canonicalized via
   1.558 +    gethostbyaddr() instead of by gethostbyname().  If gethostbyaddr()
   1.559 +    fails the gethostbyname() canonicalization is used.
   1.560 +
   1.561 +   This represents an additional security bug due to DNS spoofing, over and
   1.562 +    above trust-dns.  It also adds an additional DNS query to starting a
   1.563 +    session.
   1.564 +
   1.565 +   It is necessary for sites which implement a server cluster with multiple
   1.566 +    A records for a cluster name (instead of a CNAME) but each cluster
   1.567 +    member has a unique PTR record which it expects for a Kerberos service
   1.568 +    principal.
   1.569 +
   1.570 +   If sasl-uses-ptr-name is set to zero and trust-dns is set non-zero, the
   1.571 +    gethostbyname() canonicalized name is used for SASL authentication.
   1.572 +
   1.573 +   The setting of sasl-uses-ptr-name is irrelevant if trust-dns is set to
   1.574 +    zero.
   1.575 +
   1.576 +   The default is non-zero (use name from PTR record for SASL).
   1.577 +
   1.578 +38) set network-filesystem-stat-bug <number>
   1.579 +   By default, traditional UNIX mailbox files are only closed and reopened
   1.580 +    at checkpoint and expunge time.  This ensures that, prior to rewriting
   1.581 +    the file, that any cached stat() data from a network filesystem is
   1.582 +    updated with current data.
   1.583 +
   1.584 +   Very old versions of NFS, and reputedly also AFS, can get into a state
   1.585 +    in which the cached stat() data stays out-of-date, even across a
   1.586 +    close and reopen of the file.
   1.587 +
   1.588 +   If network-filesystem-stat-bug is set non-zero, then the mailbox file
   1.589 +    is closed and reopened at ping time as a workaround for this bug in
   1.590 +    these network filesystems.  This means that in imapd, the mailbox
   1.591 +    file is closed and reopened for every IMAP command.  This is obviously
   1.592 +    something that should be avoided unless absolutely necessary.
   1.593 +
   1.594 +   NFS and AFS are terrible ways to distribute mail.  You use use IMAP
   1.595 +    servers with a local disk instead.
   1.596 +
   1.597 +   The default is zero (only close/reopen at checkpoint and expunge time).
   1.598 +
   1.599 +   Setting this option is a great way to ruin your system's performance.
   1.600 +
   1.601 +39) set restrict-mailbox-access <option> <option> ... <option>
   1.602 +   This option is for closed server systems only.  It is less extreme
   1.603 +   than chroot-server, and allows selective restriction of what mailbox
   1.604 +   named users can use.  The existing options are:
   1.605 +    root	access not permitted to names starting with "/"
   1.606 +    otherusers	access not permitted to other users' names; this should
   1.607 +		 normally be used in conjunction with "root", otherwise
   1.608 +		 another user's names can be accessed via a root name.
   1.609 +    all		all of the above
   1.610 +   Setting any combination of options also disables access to superior
   1.611 +   directories via "..".
   1.612 +
   1.613 +   This should be done ONLY on systems which do not permit users to
   1.614 +   have shell access
   1.615 +
   1.616 +   The default is no restrictions.