README revision 303975
1# Copyright (c) 1998-2004 Proofpoint, Inc. and its suppliers. 2# All rights reserved. 3# Copyright (c) 1983, 1995-1997 Eric P. Allman. All rights reserved. 4# Copyright (c) 1988 5# The Regents of the University of California. All rights reserved. 6# 7# By using this file, you agree to the terms and conditions set 8# forth in the LICENSE file which can be found at the top level of 9# the sendmail distribution. 10# 11# 12# $Id: README,v 8.393 2013-11-22 20:51:54 ca Exp $ 13# 14 15This directory contains the source files for sendmail(TM). 16 17 ******************************************************************* 18 !! Read sendmail/SECURITY for important installation information !! 19 ******************************************************************* 20 21 ********************************************************** 22 ** Read below for more details on building sendmail. ** 23 ********************************************************** 24 25************************************************************************** 26** IMPORTANT: Read the appropriate paragraphs in the section on ** 27** ``Operating System and Compile Quirks''. ** 28************************************************************************** 29 30For detailed instructions, please read the document ../doc/op/op.me: 31 32 cd ../doc/op ; make op.ps op.txt 33 34Sendmail is a trademark of Proofpoint, Inc. 35US Patent Numbers 6865671, 6986037. 36 37 38+-------------------+ 39| BUILDING SENDMAIL | 40+-------------------+ 41 42By far, the easiest way to compile sendmail is to use the "Build" 43script: 44 45 sh ./Build 46 47This uses the "uname" command to figure out what architecture you are 48on and creates a proper Makefile accordingly. It also creates a 49subdirectory per object format, so that multiarchitecture support is 50easy. In general this should be all you need. IRIX 6.x users should 51read the note below in the OPERATING SYSTEM AND COMPILE QUIRKS section. 52 53If you need to look at other include or library directories, use the 54-I or -L flags on the command line, e.g., 55 56 sh ./Build -I/usr/sww/include -L/usr/sww/lib 57 58It's also possible to create local site configuration in the file 59site.config.m4 (or another file settable with the -f flag). This 60file contains M4 definitions for various compilation values; the 61most useful are: 62 63confMAPDEF -D flags to specify database types to be included 64 (see below) 65confENVDEF -D flags to specify other environment information 66confINCDIRS -I flags for finding include files during compilation 67confLIBDIRS -L flags for finding libraries during linking 68confLIBS -l flags for selecting libraries during linking 69confLDOPTS other ld(1) linker options 70 71Others can be found by examining Makefile.m4. Please read 72../devtools/README for more information about the site.config.m4 73file. 74 75You can recompile from scratch using the -c flag with the Build 76command. This removes the existing compilation directory for the 77current platform and builds a new one. The -c flag must also 78be used if any site.*.m4 file in devtools/Site/ is changed. 79 80Porting to a new Unix-based system should be a matter of creating 81an appropriate configuration file in the devtools/OS/ directory. 82 83 84+----------------------+ 85| DATABASE DEFINITIONS | 86+----------------------+ 87 88There are several database formats that can be used for the alias files 89and for general maps. When used for alias files they interact in an 90attempt to be backward compatible. 91 92The options are: 93 94NEWDB The new Berkeley DB package. Some systems (e.g., BSD/OS and 95 Digital UNIX 4.0) have some version of this package 96 pre-installed. If your system does not have Berkeley DB 97 pre-installed, or the version installed is not version 2.0 98 or greater (e.g., is Berkeley DB 1.85 or 1.86), get the 99 current version from http://www.sleepycat.com/. DO NOT 100 use a version from any of the University of California, 101 Berkeley "Net" or other distributions. If you are still 102 running BSD/386 1.x, you will need to upgrade the included 103 Berkeley DB library to a current version. NEWDB is included 104 automatically if the Build script can find a library named 105 libdb.a or libdb.so. 106 See also OPERATING SYSTEM AND COMPILE QUIRKS about Berkeley 107 DB versions, e.g., DB 4.1.x. 108NDBM The older NDBM implementation -- the very old V7 DBM 109 implementation is no longer supported. 110NIS Network Information Services. To use this you must have 111 NIS support on your system. 112NISPLUS NIS+ (the revised NIS released with Solaris 2). You must 113 have NIS+ support on your system to use this flag. 114HESIOD Support for Hesiod (from the DEC/Athena distribution). You 115 must already have Hesiod support on your system for this to 116 work. You may be able to get this to work with the MIT/Athena 117 version of Hesiod, but that's likely to be a lot of work. 118 BIND 8.X also includes Hesiod support. 119LDAPMAP Lightweight Directory Access Protocol support. You will 120 have to install the UMich or OpenLDAP 121 (http://www.openldap.org/) ldap and lber libraries to use 122 this flag. 123MAP_REGEX Regular Expression support. You will need to use an 124 operating system which comes with the POSIX regex() 125 routines or install a regexp library such as libregex from 126 the Free Software Foundation. 127DNSMAP DNS map support. Requires NAMED_BIND. 128PH_MAP PH map support. You will need the libphclient library from 129 the nph package (http://www-dev.cites.uiuc.edu/ph/nph/). 130MAP_NSD nsd map support (IRIX 6.5 and later). 131SOCKETMAP Support for a trivial query protocol over UNIX domain or TCP 132 sockets. 133 134>>> NOTE WELL for NEWDB support: If you want to get ndbm support, for 135>>> Berkeley DB versions under 2.0, it is CRITICAL that you remove 136>>> ndbm.o from libdb.a before you install it and DO NOT install ndbm.h; 137>>> for Berkeley DB versions 2.0 through 2.3.14, remove dbm.o from libdb.a 138>>> before you install it. If you don't delete these, there is absolutely 139>>> no point to including -DNDBM, since it will just get you another 140>>> (inferior) API to the same format database. These files OVERRIDE 141>>> calls to ndbm routines -- in particular, if you leave ndbm.h in, 142>>> you can find yourself using the new db package even if you don't 143>>> define NEWDB. Berkeley DB versions later than 2.3.14 do not need 144>>> to be modified. Please also consult the README in the top level 145>>> directory of the sendmail distribution for other important information. 146>>> 147>>> Further note: DO NOT remove your existing /usr/include/ndbm.h -- 148>>> you need that one. But do not install an updated ndbm.h in 149>>> /usr/include, /usr/local/include, or anywhere else. 150 151If NEWDB and NDBM are defined (but not NIS), then sendmail will read 152NDBM format alias files, but the next time a newaliases is run the 153format will be converted to NEWDB; that format will be used forever 154more. This is intended as a transition feature. 155 156If NEWDB, NDBM, and NIS are all defined and the name of the file includes 157the string "/yp/", sendmail will rebuild BOTH the NEWDB and NDBM format 158alias files. However, it will only read the NEWDB file; the NDBM format 159file is used only by the NIS subsystem. This is needed because the NIS 160maps on an NIS server are built directly from the NDBM files. 161 162If NDBM and NIS are defined (regardless of the definition of NEWDB), 163and the filename includes the string "/yp/", sendmail adds the special 164tokens "YP_LAST_MODIFIED" and "YP_MASTER_NAME", both of which are 165required if the NDBM file is to be used as an NIS map. 166 167All of these flags are normally defined in a confMAPDEF setting in your 168site.config.m4. 169 170If you define NEWDB or HESIOD you get the User Database (USERDB) 171automatically. Generally you do want to have NEWDB for it to do 172anything interesting. See above for getting the Berkeley DB 173package (i.e., NEWDB). There is no separate "user database" 174package -- don't bother searching for it on the net. 175 176Hesiod and LDAP require libraries that may not be installed with your 177system. These are outside of my ability to provide support. See the 178"Quirks" section for more information. 179 180The regex map can be used to see if an address matches a certain regular 181expression. For example, all-numerics local parts are common spam 182addresses, so "^[0-9]+$" would match this. By using such a map in a 183check_* rule-set, you can block a certain range of addresses that would 184otherwise be considered valid. 185 186The socket map uses a simple request/reply protocol over TCP or 187UNIX domain sockets to query an external server. Both requests and 188replies are text based and encoded as netstrings. The socket map 189uses the same syntax as milters the specify the remote endpoint, 190e.g.: 191 192KmySocketMap socket inet:12345@127.0.0.1 193 194See doc/op/op.me for details. 195 196 197+---------------+ 198| COMPILE FLAGS | 199+---------------+ 200 201Wherever possible, I try to make sendmail pull in the correct 202compilation options needed to compile on various environments based on 203automatically defined symbols. Some machines don't seem to have useful 204symbols available, requiring that a compilation flag be defined in 205the Makefile; see the devtools/OS subdirectory for the supported 206architectures. 207 208If you are a system to which sendmail has already been ported you 209should not have to touch the following symbols. But if you are porting, 210you may have to tweak the following compilation flags in conf.h in order 211to get it to compile and link properly: 212 213SYSTEM5 Adjust for System V (not necessarily Release 4). 214SYS5SIGNALS Use System V signal semantics -- the signal handler 215 is automatically dropped when the signal is caught. 216 If this is not set, use POSIX/BSD semantics, where the 217 signal handler stays in force until an exec or an 218 explicit delete. Implied by SYSTEM5. 219SYS5SETPGRP Use System V setpgrp() semantics. Implied by SYSTEM5. 220HASNICE Define this to zero if you lack the nice(2) system call. 221HASRRESVPORT Define this to zero if you lack the rresvport(3) system call. 222HASFCHMOD Define this to one if you have the fchmod(2) system call. 223 This improves security. 224HASFCHOWN Define this to one if you have the fchown(2) system call. 225 This is required for the TrustedUser option if sendmail 226 must rebuild an (alias) map. 227HASFLOCK Set this if you prefer to use the flock(2) system call 228 rather than using fcntl-based locking. Fcntl locking 229 has some semantic gotchas, but many vendor systems 230 also interface it to lockd(8) to do NFS-style locking. 231 Unfortunately, may vendors implementations of fcntl locking 232 is just plain broken (e.g., locks are never released, 233 causing your sendmail to deadlock; when the kernel runs 234 out of locks your system crashes). For this reason, I 235 recommend always defining this unless you are absolutely 236 certain that your fcntl locking implementation really works. 237HASUNAME Set if you have the "uname" system call. Implied by 238 SYSTEM5. 239HASUNSETENV Define this if your system library has the "unsetenv" 240 subroutine. 241HASSETSID Define this if you have the setsid(2) system call. This 242 is implied if your system appears to be POSIX compliant. 243HASINITGROUPS Define this if you have the initgroups(3) routine. 244HASSETVBUF Define this if you have the setvbuf(3) library call. 245 If you don't, setlinebuf will be used instead. This 246 defaults on if your compiler defines __STDC__. 247HASSETREUID Define this if you have setreuid(2) ***AND*** root can 248 use setreuid to change to an arbitrary user. This second 249 condition is not satisfied on AIX 3.x. You may find that 250 your system has setresuid(2), (for example, on HP-UX) in 251 which case you will also have to #define setreuid(r, e) 252 to be the appropriate call. Some systems (such as Solaris) 253 have a compatibility routine that doesn't work properly, 254 but may have "saved user ids" properly implemented so you 255 can ``#define setreuid(r, e) seteuid(e)'' and have it work. 256 The important thing is that you have a call that will set 257 the effective uid independently of the real or saved uid 258 and be able to set the effective uid back again when done. 259 There's a test program in ../test/t_setreuid.c that will 260 try things on your system. Setting this improves the 261 security, since sendmail doesn't have to read .forward 262 and :include: files as root. There are certain attacks 263 that may be unpreventable without this call. 264USESETEUID Define this to 1 if you have a seteuid(2) system call that 265 will allow root to set only the effective user id to an 266 arbitrary value ***AND*** you have saved user ids. This is 267 preferable to HASSETREUID if these conditions are fulfilled. 268 These are the semantics of the to-be-released revision of 269 Posix.1. The test program ../test/t_seteuid.c will try 270 this out on your system. If you define both HASSETREUID 271 and USESETEUID, the former is ignored. 272HASSETEGID Define this if you have setegid(2) and it can be 273 used to set the saved gid. Please run t_dropgid in 274 test/ if you are not sure whether the call works. 275HASSETREGID Define this if you have setregid(2) and it can be 276 used to set the saved gid. Please run t_dropgid in 277 test/ if you are not sure whether the call works. 278HASSETRESGID Define this if you have setresgid(2) and it can be 279 used to set the saved gid. Please run t_dropgid in 280 test/ if you are not sure whether the call works. 281HASLSTAT Define this if you have symbolic links (and thus the 282 lstat(2) system call). This improves security. Unlike 283 most other options, this one is on by default, so you 284 need to #undef it in conf.h if you don't have symbolic 285 links (these days everyone does). 286HASSETRLIMIT Define this to 1 if you have the setrlimit(2) syscall. 287 You can define it to 0 to force it off. It is assumed 288 if you are running a BSD-like system. 289HASULIMIT Define this if you have the ulimit(2) syscall (System V 290 style systems). HASSETRLIMIT overrides, as it is more 291 general. 292HASWAITPID Define this if you have the waitpid(2) syscall. 293HASGETDTABLESIZE 294 Define this if you have the getdtablesize(2) syscall. 295HAS_ST_GEN Define this to 1 if your system has the st_gen field in 296 the stat structure (see stat(2)). 297HASSRANDOMDEV Define this if your system has the srandomdev(3) function 298 call. 299HASURANDOMDEV Define this if your system has /dev/urandom(4). 300HASSTRERROR Define this if you have the libc strerror(3) function (which 301 should be declared in <errno.h>), and it should be used 302 instead of sys_errlist. 303HASCLOSEFROM Define this if your system has closefrom(3). 304HASFDWALK Define this if your system has fdwalk(3). 305SM_CONF_GETOPT Define this as 0 if you need a reimplementation of getopt(3). 306 On some systems, getopt does very odd things if called 307 to scan the arguments twice. This flag will ask sendmail 308 to compile in a local version of getopt that works 309 properly. You may also need this if you build with 310 another library that introduces a non-standard getopt(3). 311NEEDSTRTOL Define this if your standard C library does not define 312 strtol(3). This will compile in a local version. 313NEEDFSYNC Define this if your standard C library does not define 314 fsync(2). This will try to simulate the operation using 315 fcntl(2); if that is not available it does nothing, which 316 isn't great, but at least it compiles and runs. 317HASGETUSERSHELL Define this to 1 if you have getusershell(3) in your 318 standard C library. If this is not defined, or is defined 319 to be 0, sendmail will scan the /etc/shells file (no 320 NIS-style support, defaults to /bin/sh and /bin/csh if 321 that file does not exist) to get a list of unrestricted 322 user shells. This is used to determine whether users 323 are allowed to forward their mail to a program or a file. 324NEEDPUTENV Define this if your system needs am emulation of the 325 putenv(3) call. Define to 1 to implement it in terms 326 of setenv(3) or to 2 to do it in terms of primitives. 327NOFTRUNCATE Define this if you don't have the ftruncate(2) syscall. 328 If you don't have this system call, there is an unavoidable 329 race condition that occurs when creating alias databases. 330GIDSET_T The type of entries in a gidset passed as the second 331 argument to getgroups(2). Historically this has been an 332 int, so this is the default, but some systems (such as 333 IRIX) pass it as a gid_t, which is an unsigned short. 334 This will make a difference, so it is important to get 335 this right! However, it is only an issue if you have 336 group sets. 337SLEEP_T The type returned by the system sleep() function. 338 Defaults to "unsigned int". Don't worry about this 339 if you don't have compilation problems. 340ARBPTR_T The type of an arbitrary pointer -- defaults to "void *". 341 If you are an very old compiler you may need to define 342 this to be "char *". 343SOCKADDR_LEN_T The type used for the third parameter to accept(2), 344 getsockname(2), and getpeername(2), representing the 345 length of a struct sockaddr. Defaults to int. 346SOCKOPT_LEN_T The type used for the fifth parameter to getsockopt(2) 347 and setsockopt(2), representing the length of the option 348 buffer. Defaults to int. 349LA_TYPE The type of load average your kernel supports. These 350 can be one of: 351 LA_ZERO (1) -- it always returns the load average as 352 "zero" (and does so on all architectures). 353 LA_INT (2) to read /dev/kmem for the symbol avenrun and 354 interpret as a long integer. 355 LA_FLOAT (3) same, but interpret the result as a floating 356 point number. 357 LA_SHORT (6) to interpret as a short integer. 358 LA_SUBR (4) if you have the getloadavg(3) routine in your 359 system library. 360 LA_MACH (5) to use MACH-style load averages (calls 361 processor_set_info()), 362 LA_PROCSTR (7) to read /proc/loadavg and interpret it 363 as a string representing a floating-point 364 number (Linux-style). 365 LA_READKSYM (8) is an implementation suitable for some 366 versions of SVr4 that uses the MIOC_READKSYM ioctl 367 call to read /dev/kmem. 368 LA_DGUX (9) is a special implementation for DG/UX that uses 369 the dg_sys_info system call. 370 LA_HPUX (10) is an HP-UX specific version that uses the 371 pstat_getdynamic system call. 372 LA_IRIX6 (11) is an IRIX 6.x specific version that adapts 373 to 32 or 64 bit kernels; it is otherwise very similar 374 to LA_INT. 375 LA_KSTAT (12) uses the (Solaris-specific) kstat(3k) 376 implementation. 377 LA_DEVSHORT (13) reads a short from a system file (default: 378 /dev/table/avenrun) and scales it in the same manner 379 as LA_SHORT. 380 LA_LONGLONG (17) to read /dev/kmem for the symbol avenrun and 381 interpret as a long long integer (e.g., for 64 bit 382 systems). 383 LA_INT, LA_SHORT, LA_FLOAT, and LA_READKSYM have several 384 other parameters that they try to divine: the name of your 385 kernel, the name of the variable in the kernel to examine, 386 the number of bits of precision in a fixed point load average, 387 and so forth. LA_DEVSHORT uses _PATH_AVENRUN to find the 388 device to be read to find the load average. 389 In desperation, use LA_ZERO. The actual code is in 390 conf.c -- it can be tweaked if you are brave. 391FSHIFT For LA_INT, LA_SHORT, and LA_READKSYM, this is the number 392 of bits of load average after the binary point -- i.e., 393 the number of bits to shift right in order to scale the 394 integer to get the true integer load average. Defaults to 8. 395_PATH_UNIX The path to your kernel. Needed only for LA_INT, LA_SHORT, 396 and LA_FLOAT. Defaults to "/unix" on System V, "/vmunix" 397 everywhere else. 398LA_AVENRUN For LA_INT, LA_SHORT, and LA_FLOAT, the name of the kernel 399 variable that holds the load average. Defaults to "avenrun" 400 on System V, "_avenrun" everywhere else. 401SFS_TYPE Encodes how your kernel can locate the amount of free 402 space on a disk partition. This can be set to SFS_NONE 403 (0) if you have no way of getting this information, 404 SFS_USTAT (1) if you have the ustat(2) system call, 405 SFS_4ARGS (2) if you have a four-argument statfs(2) 406 system call (and the include file is <sys/statfs.h>), 407 SFS_VFS (3), SFS_MOUNT (4), SFS_STATFS (5) if you have 408 the two-argument statfs(2) system call with includes in 409 <sys/vfs.h>, <sys/mount.h>, or <sys/statfs.h> respectively, 410 or SFS_STATVFS (6) if you have the two-argument statvfs(2) 411 call. The default if nothing is defined is SFS_NONE. 412SFS_BAVAIL with SFS_4ARGS you can also set SFS_BAVAIL to the field name 413 in the statfs structure that holds the useful information; 414 this defaults to f_bavail. 415SPT_TYPE Encodes how your system can display what a process is doing 416 on a ps(1) command (SPT stands for Set Process Title). Can 417 be set to: 418 SPT_NONE (0) -- Don't try to set the process title at all. 419 SPT_REUSEARGV (1) -- Pad out your argv with the information; 420 this is the default if none specified. 421 SPT_BUILTIN (2) -- The system library has setproctitle. 422 SPT_PSTAT (3) -- Use the PSTAT_SETCMD option to pstat(2) 423 to set the process title; this is used by HP-UX. 424 SPT_PSSTRINGS (4) -- Use the magic PS_STRINGS pointer (4.4BSD). 425 SPT_SYSMIPS (5) -- Use sysmips() supported by NEWS-OS 6. 426 SPT_SCO (6) -- Write kernel u. area. 427 SPT_CHANGEARGV (7) -- Write pointers to our own strings into 428 the existing argv vector. 429SPT_PADCHAR Character used to pad the process title; if undefined, 430 the space character (0x20) is used. This is ignored if 431 SPT_TYPE != SPT_REUSEARGV 432ERRLIST_PREDEFINED 433 If set, assumes that some header file defines sys_errlist. 434 This may be needed if you get type conflicts on this 435 variable -- otherwise don't worry about it. 436WAITUNION The wait(2) routine takes a "union wait" argument instead 437 of an integer argument. This is for compatibility with 438 old versions of BSD. 439SCANF You can set this to extend the F command to accept a 440 scanf string -- this gives you a primitive parser for 441 class definitions -- BUT it can make you vulnerable to 442 core dumps if the target file is poorly formed. 443SYSLOG_BUFSIZE You can define this to be the size of the buffer that 444 syslog accepts. If it is not defined, it assumes a 445 1024-byte buffer. If the buffer is very small (under 446 256 bytes) the log message format changes -- each 447 e-mail message will log many more messages, since it 448 will log each piece of information as a separate line 449 in syslog. 450BROKEN_RES_SEARCH 451 On Ultrix (and maybe other systems?) if you use the 452 res_search routine with an unknown host name, it returns 453 -1 but sets h_errno to 0 instead of HOST_NOT_FOUND. If 454 you set this, sendmail considers 0 to be the same as 455 HOST_NOT_FOUND. 456NAMELISTMASK If defined, values returned by nlist(3) are masked 457 against this value before use -- a common value is 458 0x7fffffff to strip off the top bit. 459BSD4_4_SOCKADDR If defined, socket addresses have an sa_len field that 460 defines the length of this address. 461SAFENFSPATHCONF Set this to 1 if and only if you have verified that a 462 pathconf(2) call with _PC_CHOWN_RESTRICTED argument on an 463 NFS filesystem where the underlying system allows users to 464 give away files to other users returns <= 0. Be sure you 465 try both on NFS V2 and V3. Some systems assume that their 466 local policy apply to NFS servers -- this is a bad 467 assumption! The test/t_pathconf.c program will try this 468 for you -- you have to run it in a directory that is 469 mounted from a server that allows file giveaway. 470SIOCGIFCONF_IS_BROKEN 471 Set this if your system has an SIOCGIFCONF ioctl defined, 472 but it doesn't behave the same way as "most" systems (BSD, 473 Solaris, SunOS, HP-UX, etc.) 474SIOCGIFNUM_IS_BROKEN 475 Set this if your system has an SIOCGIFNUM ioctl defined, 476 but it doesn't behave the same way as "most" systems 477 (Solaris, HP-UX). 478FAST_PID_RECYCLE 479 Set this if your system can reuse the same PID in the same 480 second. 481SO_REUSEADDR_IS_BROKEN 482 Set this if your system has a setsockopt() SO_REUSEADDR 483 flag but doesn't pay attention to it when trying to bind a 484 socket to a recently closed port. 485NEEDSGETIPNODE Set this if your system supports IPv6 but doesn't include 486 the getipnodeby{name,addr}() functions. Set automatically 487 for Linux's glibc. 488PIPELINING Support SMTP PIPELINING (set by default). 489USING_NETSCAPE_LDAP 490 Deprecated in favor of SM_CONF_LDAP_MEMFREE. See 491 libsm/README. 492NEEDLINK Set this if your system doesn't have a link() call. It 493 will create a copy of the file instead of a hardlink. 494USE_ENVIRON Set this to 1 to access process environment variables from 495 the external variable environ instead of the third 496 parameter of main(). 497USE_DOUBLE_FORK By default this is on (1). Set it to 0 to suppress the 498 extra fork() used to avoid intermediate zombies. 499ALLOW_255 Do not convert (char)0xff to (char)0x7f in headers etc. 500 This can also be done at runtime with the command line 501 option -d82.101. 502NEEDINTERRNO Set this if <errno.h> does not declare errno, i.e., if an 503 application needs to use 504 extern int errno; 505USE_TTYPATH Set this to 1 to enable ErrorMode=write. 506USESYSCTL Use sysctl(3) to determine the number of CPUs in a system. 507HASSNPRINTF Set this to 1 if your OS has a working snprintf(3), i.e., 508 it properly obeys the size of the buffer and returns the 509 number of characters that would have been printed if the 510 size were unlimited. 511LDAP_REFERRALS Set this if you want to use the -R flag (do not auto chase 512 referrals) for LDAP maps (requires -DLDAPMAP). 513MILTER_NO_NAGLE Turn off Nagle algorithm for communication with libmilter 514 ("cork" on Linux). On some operating systems this may 515 improve the interprocess communication performance. 516 517 518+-----------------------+ 519| COMPILE-TIME FEATURES | 520+-----------------------+ 521 522There are a bunch of features that you can decide to compile in, such 523as selecting various database packages and special protocol support. 524Several are assumed based on other compilation flags -- if you want to 525"un-assume" something, you probably need to edit conf.h. Compilation 526flags that add support for special features include: 527 528NDBM Include support for "new" DBM library for aliases and maps. 529 Normally defined in the Makefile. 530NEWDB Include support for Berkeley DB package (hash & btree) 531 for aliases and maps. Normally defined in the Makefile. 532 If the version of NEWDB you have is the old one that does 533 not include the "fd" call (this call was added in version 534 1.5 of the Berkeley DB code), you must upgrade to the 535 current version of Berkeley DB. 536NIS Define this to get NIS (YP) support for aliases and maps. 537 Normally defined in the Makefile. 538NISPLUS Define this to get NIS+ support for aliases and maps. 539 Normally defined in the Makefile. 540HESIOD Define this to get Hesiod support for aliases and maps. 541 Normally defined in the Makefile. 542NETINFO Define this to get NeXT NetInfo support for aliases and maps. 543 Normally defined in the Makefile. 544LDAPMAP Define this to get LDAP support for maps. 545PH_MAP Define this to get PH support for maps. 546MAP_NSD Define this to get nsd support for maps. 547USERDB Define this to 1 to include support for the User Information 548 Database. Implied by NEWDB or HESIOD. You can use 549 -DUSERDB=0 to explicitly turn it off. 550IDENTPROTO Define this as 1 to get IDENT (RFC 1413) protocol support. 551 This is assumed unless you are running on Ultrix or 552 HP-UX, both of which have a problem in the UDP 553 implementation. You can define it to be 0 to explicitly 554 turn off IDENT protocol support. If defined off, the code 555 is actually still compiled in, but it defaults off; you 556 can turn it on by setting the IDENT timeout in the 557 configuration file. 558IP_SRCROUTE Define this to 1 to get IP source routing information 559 displayed in the Received: header. This is assumed on 560 most systems, but some (e.g., Ultrix) apparently have a 561 broken version of getsockopt that doesn't properly 562 support the IP_OPTIONS call. You probably want this if 563 your OS can cope with it. Symptoms of failure will be that 564 it won't compile properly (that is, no support for fetching 565 IP_OPTIONs), or it compiles but source-routed TCP connections 566 either refuse to open or open and hang for no apparent reason. 567 Ultrix and AIX3 are known to fail this way. 568LOG Set this to get syslog(3) support. Defined by default 569 in conf.h. You want this if at all possible. 570NETINET Set this to get TCP/IP support. Defined by default 571 in conf.h. You probably want this. 572NETINET6 Set this to get IPv6 support. Other configuration may 573 be needed in conf.h for your particular operating system. 574 Also, DaemonPortOptions must be set appropriately for 575 sendmail to accept IPv6 connections. 576NETISO Define this to get ISO networking support. 577NETUNIX Define this to get Unix domain networking support. Defined 578 by default. A few bizarre systems (SCO, ISC, Altos) don't 579 support this networking domain. 580NETNS Define this to get NS networking support. 581NETX25 Define this to get X.25 networking support. 582NAMED_BIND If non-zero, include DNS (name daemon) support, including 583 MX support. The specs say you must use this if you run 584 SMTP. You don't have to be running a name server daemon 585 on your machine to need this -- any use of the DNS resolver, 586 including remote access to another machine, requires this 587 option. Defined by default in conf.h. Define it to zero 588 ONLY on machines that do not use DNS in any way. 589MATCHGECOS Permit fuzzy matching of user names against the full 590 name (GECOS) field in the /etc/passwd file. This should 591 probably be on, since you can disable it from the config 592 file if you want to. Defined by default in conf.h. 593MIME8TO7 If non-zero, include 8 to 7 bit MIME conversions. This 594 also controls advertisement of 8BITMIME in the ESMTP 595 startup dialogue. 596MIME7TO8_OLD If 0 then use an algorithm for MIME 7-bit quoted-printable 597 or base64 encoding to 8-bit text that has been introduced 598 in 8.12.3. There are some examples where that code fails, 599 but the old code works. If you have an example of improper 600 7 to 8 bit conversion please send it to sendmail-bugs. 601MIME7TO8 If non-zero, include 7 to 8 bit MIME conversions. 602HES_GETMAILHOST Define this to 1 if you are using Hesiod with the 603 hes_getmailhost() routine. This is included with the MIT 604 Hesiod distribution, but not with the DEC Hesiod distribution. 605XDEBUG Do additional internal checking. These don't cost too 606 much; you might as well leave this on. 607TCPWRAPPERS Turns on support for the TCP wrappers library (-lwrap). 608 See below for further information. 609SECUREWARE Enable calls to the SecureWare luid enabling/changing routines. 610 SecureWare is a C2 security package added to several UNIX's 611 (notably ConvexOS) to get a C2 Secure system. This 612 option causes mail delivery to be done with the luid of the 613 recipient. 614SHARE_V1 Support for the fair share scheduler, version 1. Setting to 615 1 causes final delivery to be done using the recipients 616 resource limitations. So far as I know, this is only 617 supported on ConvexOS. 618SASL Enables SMTP AUTH (RFC 2554). This requires the Cyrus SASL 619 library (ftp://ftp.andrew.cmu.edu/pub/cyrus-mail/). Please 620 install at least version 1.5.13. See below for further 621 information: SASL COMPILATION AND CONFIGURATION. If your 622 SASL library is older than 1.5.10, you have to set this 623 to its version number using a simple conversion: a.b.c 624 -> c + b*100 + a*10000, e.g. for 1.5.9 define SASL=10509. 625 Note: Using an older version than 1.5.5 of Cyrus SASL is 626 not supported. Starting with version 1.5.10, setting SASL=1 627 is sufficient. Any value other than 1 (or 0) will be 628 compared with the actual version found and if there is a 629 mismatch, compilation will fail. 630EGD Define this if your system has EGD installed, see 631 http://egd.sourceforge.net/ . It should be used to 632 seed the PRNG for STARTTLS if HASURANDOMDEV is not defined. 633STARTTLS Enables SMTP STARTTLS (RFC 2487). This requires OpenSSL 634 (http://www.OpenSSL.org/); use OpenSSL 0.9.8zc or later. 635 See STARTTLS COMPILATION AND CONFIGURATION for further 636 information. 637TLS_NO_RSA Turn off support for RSA algorithms in STARTTLS. 638MILTER Turn on support for external filters using the Milter API; 639 this option is set by default, to turn it off use 640 APPENDDEF(`conf_sendmail_ENVDEF', `-DMILTER=0') 641 in devtools/Site/site.config.m4 (see devtools/README). 642 See libmilter/README for more information about milter. 643REQUIRES_DIR_FSYNC Turn on support for file systems that require to 644 call fsync() for a directory if the meta-data in it has 645 been changed. This should be turned on at least for older 646 versions of ReiserFS; it is enabled by default for Linux. 647 According to some information this flag is not needed 648 anymore for kernel 2.4.16 and newer. We would appreciate 649 feedback about the semantics of the various file systems 650 available for Linux. 651 An alternative to this compile time flag is to mount the 652 queue directory without the -async option, or using 653 chattr +S on Linux. 654DBMMODE The default file permissions to use when creating new 655 database files for maps and aliases. Defaults to 0640. 656IPV6_FULL Use uncompressed IPv6 addresses (set by default). This 657 permits a zero subnet to have a more specific match, 658 such as different map entries for IPv6:0:0 vs IPv6:0. 659 660Generic notice: If you enable a compile time option that needs 661libraries or include files that don't come with sendmail or are 662installed in a location that your C compiler doesn't use by default 663you should set confINCDIRS and confLIBDIRS as explained in the 664first section: BUILDING SENDMAIL. 665 666 667+---------------------+ 668| DNS/RESOLVER ISSUES | 669+---------------------+ 670 671Many systems have old versions of the resolver library. At a minimum, 672you should be running BIND 4.8.3; older versions may compile, but they 673have known bugs that should give you pause. 674 675Common problems in old versions include "undefined" errors for 676dn_skipname. 677 678Some people have had a problem with BIND 4.9; it uses some routines 679that it expects to be externally defined such as strerror(). It may 680help to link with "-l44bsd" to solve this problem. This has apparently 681been fixed in later versions of BIND, starting around 4.9.3. In other 682words, if you use 4.9.0 through 4.9.2, you need -l44bsd; for earlier or 683later versions, you do not. 684 685!PLEASE! be sure to link with the same version of the resolver as 686the header files you used -- some people have used the 4.9 headers 687and linked with BIND 4.8 or vice versa, and it doesn't work. 688Unfortunately, it doesn't fail in an obvious way -- things just 689subtly don't work. 690 691WILDCARD MX RECORDS ARE A BAD IDEA! The only situation in which they 692work reliably is if you have two versions of DNS, one in the real world 693which has a wildcard pointing to your firewall, and a completely 694different version of the database internally that does not include 695wildcard MX records that match your domain. ANYTHING ELSE WILL GIVE 696YOU HEADACHES! 697 698When attempting to canonify a hostname, some broken name servers will 699return SERVFAIL (a temporary failure) on T_AAAA (IPv6) lookups. If you 700want to excuse this behavior, include WorkAroundBrokenAAAA in 701ResolverOptions. However, instead, we recommend catching the problem and 702reporting it to the name server administrator so we can rid the world of 703broken name servers. 704 705 706+----------------------------------------+ 707| STARTTLS COMPILATION AND CONFIGURATION | 708+----------------------------------------+ 709 710Please read the documentation accompanying the OpenSSL library. You 711have to compile and install the OpenSSL libraries before you can compile 712sendmail. See devtools/README how to set the correct compile time 713parameters; you should at least set the following variables: 714 715APPENDDEF(`conf_sendmail_ENVDEF', `-DSTARTTLS') 716APPENDDEF(`conf_sendmail_LIBS', `-lssl -lcrypto') 717 718If you have installed the OpenSSL libraries and include files in 719a location that your C compiler doesn't use by default you should 720set confINCDIRS and confLIBDIRS as explained in the first section: 721BUILDING SENDMAIL. 722 723Configuration information can be found in doc/op/op.me (required 724certificates) and cf/README (how to tell sendmail about certificates). 725 726To perform an initial test, connect to your sendmail daemon 727(telnet localhost 25) and issue a EHLO localhost and see whether 728250-STARTTLS 729is in the response. If it isn't, run the daemon with 730-O LogLevel=14 731and try again. Then take a look at the logfile and see whether 732there are any problems listed about permissions (unsafe files) 733or the validity of X.509 certificates. 734 735From: Garrett Wollman <wollman@lcs.mit.edu> 736 737 If your certificate authority is hierarchical, and you only include 738 the top-level CA certificate in the CACertFile file, some mail clients 739 may be unable to infer the proper certificate chain when selecting a 740 client certificate. Including the bottom-level CA certificate(s) in 741 the CACertFile file will allow these clients to work properly. This 742 is not necessary if you are not using client certificates for 743 authentication, or if all your clients are running Sendmail or other 744 programs using the OpenSSL library (which get it right automatically). 745 In addition, some mail clients are totally incapable of using 746 certificate authentication -- even some of those which already support 747 SSL/TLS for confidentiality. 748 749Further information can be found via: 750http://www.sendmail.org/tips/ 751 752 753+------------------------------------+ 754| SASL COMPILATION AND CONFIGURATION | 755+------------------------------------+ 756 757Please read the documentation accompanying the Cyrus SASL library 758(INSTALL and README). If you use Berkeley DB for Cyrus SASL then 759you must compile sendmail with the same version of Berkeley DB. 760See devtools/README for how to set the correct compile time parameters; 761you should at least set the following variables: 762 763APPENDDEF(`conf_sendmail_ENVDEF', `-DSASL') 764APPENDDEF(`conf_sendmail_LIBS', `-lsasl') 765 766If you have installed the Cyrus SASL library and include files in 767a location that your C compiler doesn't use by default you should 768set confINCDIRS and confLIBDIRS as explained in the first section: 769BUILDING SENDMAIL. 770 771You have to select and install authentication mechanisms and tell 772sendmail where to find the sasl library and the include files (see 773devtools/README for the parameters to set). Set up the required 774users and passwords as explained in the SASL documentation. See 775also cf/README for authentication related options (especially 776DefaultAuthInfo if you want authentication between MTAs). 777 778To perform an initial test, connect to your sendmail daemon 779(telnet localhost 25) and issue a EHLO localhost and see whether 780250-AUTH .... 781is in the response. If it isn't, run the daemon with 782-O LogLevel=14 783and try again. Then take a look at the logfile and see whether 784there are any security related problems listed (unsafe files). 785 786Further information can be found via: 787http://www.sendmail.org/tips/ 788 789 790+-------------------------------------+ 791| OPERATING SYSTEM AND COMPILE QUIRKS | 792+-------------------------------------+ 793 794GCC problems 795 When compiling with "gcc -O -Wall" specify "-DSM_OMIT_BOGUS_WARNINGS" 796 too (see include/sm/cdefs.h for more info). 797 798 ***************************************************************** 799 ** IMPORTANT: DO NOT USE OPTIMIZATION (``-O'') IF YOU ARE ** 800 ** RUNNING GCC 2.4.x or 2.5.x. THERE IS A BUG IN THE GCC ** 801 ** OPTIMIZER THAT CAUSES SENDMAIL COMPILES TO FAIL MISERABLY. ** 802 ***************************************************************** 803 804 Jim Wilson of Cygnus believes he has found the problem -- it will 805 probably be fixed in GCC 2.5.6 -- but until this is verified, be 806 very suspicious of gcc -O. This problem is reported to have been 807 fixed in gcc 2.6. 808 809 A bug in gcc 2.5.5 caused problems compiling sendmail 8.6.5 with 810 optimization on a Sparc. If you are using gcc 2.5.5, youi should 811 upgrade to the latest version of gcc. 812 813 Apparently GCC 2.7.0 on the Pentium processor has optimization 814 problems. I recommend against using -O on that architecture. This 815 has been seen on FreeBSD 2.0.5 RELEASE. 816 817 Solaris 2.X users should use version 2.7.2.3 over 2.7.2. 818 819 We have been told there are problems with gcc 2.8.0. If you are 820 using this version, you should upgrade to 2.8.1 or later. 821 822Berkeley DB 823 Berkeley DB 4.1.x with x <= 24 does not work with sendmail. 824 You need at least 4.1.25. 825 826GDBM GDBM does not work with sendmail because the additional 827 security checks and file locking cause problems. Unfortunately, 828 gdbm does not provide a compile flag in its version of ndbm.h so 829 the code can adapt. Until the GDBM authors can fix these problems, 830 GDBM will not be supported. Please use Berkeley DB instead. 831 832Configuration file location 833 Up to 8.6, sendmail tried to find the sendmail.cf file in the same 834 place as the vendors had put it, even when this was obviously 835 stupid. As of 8.7, sendmail ALWAYS looks for /etc/sendmail.cf. 836 Beginning with 8.10, sendmail uses /etc/mail/sendmail.cf. 837 You can get sendmail to use the stupid vendor .cf location by 838 adding -DUSE_VENDOR_CF_PATH during compilation, but this may break 839 support programs and scripts that need to find sendmail.cf. You 840 are STRONGLY urged to use symbolic links if you want to use the 841 vendor location rather than changing the location in the sendmail 842 binary. 843 844 NETINFO systems use NETINFO to determine the location of 845 sendmail.cf. The full path to sendmail.cf is stored as the value of 846 the "sendmail.cf" property in the "/locations/sendmail" 847 subdirectory of NETINFO. Set the value of this property to 848 "/etc/mail/sendmail.cf" (without the quotes) to use this new 849 default location for Sendmail 8.10.0 and higher. 850 851ControlSocket permissions 852 Paraphrased from BIND 8.2.1's README: 853 854 Solaris and other pre-4.4BSD kernels do not respect ownership or 855 protections on UNIX-domain sockets. The short term fix for this is to 856 override the default path and put such control sockets into root- 857 owned directories which do not permit non-root to r/w/x through them. 858 The long term fix is for all kernels to upgrade to 4.4BSD semantics. 859 860HP MPE/iX 861 The MPE-specific code within sendmail emulates a set-user-id root 862 environment for the sendmail binary. But there is no root uid 0 on 863 MPE, nor is there any support for set-user-id programs. Even when 864 sendmail thinks it is running as uid 0, it will still have the file 865 access rights of the underlying non-zero uid, but because sendmail is 866 an MPE priv-mode program it will still be able to call setuid() to 867 successfully switch to a new uid. 868 869 MPE setgid() semantics don't quite work the way sendmail expects, so 870 special emulation is done here also. 871 872 This uid/gid emulation is enabled via the setuid/setgid file mode bits 873 which are not currently used by MPE. Code in libsm/mpeix.c examines 874 these bits and enables emulation if they have been set, i.e., 875 chmod u+s,g+s /SENDMAIL/CURRENT/SENDMAIL. 876 877SunOS 4.x (Solaris 1.x) 878 You may have to use -lresolv on SunOS. However, beware that 879 this links in a new version of gethostbyname that does not 880 understand NIS, so you must have all of your hosts in DNS. 881 882 Some people have reported problems with the SunOS version of 883 -lresolv and/or in.named, and suggest that you get a newer 884 version. The symptoms are delays when you connect to the 885 SMTP server on a SunOS machine or having your domain added to 886 addresses inappropriately. There is a version of BIND 887 version 4.9 on gatekeeper.DEC.COM in pub/BSD/bind/4.9. 888 889 There is substantial disagreement about whether you can make 890 this work with resolv+, which allows you to specify a search-path 891 of services. Some people report that it works fine, others 892 claim it doesn't work at all (including causing sendmail to 893 drop core when it tries to do multiple resolv+ lookups for a 894 single job). I haven't tried resolv+, as we use DNS exclusively. 895 896 Should you want to try resolv+, it is on ftp.uu.net in 897 /networking/ip/dns. 898 899 Apparently getservbyname() can fail under moderate to high 900 load under some circumstances. This will exhibit itself as 901 the message ``554 makeconnection: service "smtp" unknown''. 902 The problem has been traced to one or more blank lines in 903 /etc/services on the NIS server machine. Delete these 904 and it should work. This info is thanks to Brian Bartholomew 905 <bb@math.ufl.edu> of I-Kinetics, Inc. 906 907 NOTE: The SunOS 4.X linker uses library paths specified during 908 compilation using -L for run-time shared library searches. 909 Therefore, it is vital that relative and unsafe directory paths not 910 be used when compiling sendmail. 911 912SunOS 4.0.2 (Sun 386i) 913 Date: Fri, 25 Aug 1995 11:13:58 +0200 (MET DST) 914 From: teus@oce.nl 915 916 Sendmail 8.7.Beta.12 compiles and runs nearly out of the box with the 917 following changes: 918 * Don't use /usr/5bin in your PATH, but make /usr/5bin/uname 919 available as "uname" command. 920 * Use the defines "-DBSD4_3 -DNAMED_BIND=0" in 921 devtools/OS/SunOS.4.0, which is selected via the "uname" command. 922 I recommend to make available the db-library on the system first 923 (and change the Makefile to use this library). 924 Note that the sendmail.cf and aliases files are found in /etc. 925 926SunOS 4.1.3, 4.1.3_U1 927 Sendmail causes crashes on SunOS 4.1.3 and 4.1.3_U1. According 928 to Sun bug number 1077939: 929 930 If an application does a getsockopt() on a SOCK_STREAM (TCP) socket 931 after the other side of the connection has sent a TCP RESET for 932 the stream, the kernel gets a Bus Trap in the tcp_ctloutput() or 933 ip_ctloutput() routine. 934 935 For 4.1.3, this is fixed in patch 100584-08, available on the 936 Sunsolve 2.7.1 or later CDs. For 4.1.3_U1, this was fixed in patch 937 101790-01 (SunOS 4.1.3_U1: TCP socket and reset problems), later 938 obsoleted by patch 102010-05. 939 940 Sun patch 100584-08 is not currently publicly available on their 941 ftp site but a user has reported it can be found at other sites 942 using a web search engine. 943 944Solaris 2.x (SunOS 5.x) 945 To compile for Solaris, the Makefile built by Build must 946 include a SOLARIS definition which reflects the Solaris version 947 (i.e. -DSOLARIS=20400 for 2.4 or -DSOLARIS=20501 for 2.5.1). 948 If you are using gcc, make sure -I/usr/include is not used (or 949 it might complain about TopFrame). If you are using Sun's cc, 950 make sure /opt/SUNWspro/bin/cc is used instead of /usr/ucb/cc 951 (or it might complain about tm_zone). 952 953 The Solaris 2.x (x <= 3) "syslog" function is apparently limited 954 to something about 90 characters because of a kernel limitation. 955 If you have source code, you can probably up this number. You 956 can get patches that fix this problem: the patch ids are: 957 958 Solaris 2.1 100834 959 Solaris 2.2 100999 960 Solaris 2.3 101318 961 962 Be sure you have the appropriate patch installed or you won't 963 see system logging. 964 965Solaris 2.4 (SunOS 5.4) 966 If you include /usr/lib at the end of your LD_LIBRARY_PATH you run 967 the risk of getting the wrong libraries under some circumstances. 968 This is because of a new feature in Solaris 2.4, described by 969 Rod.Evans@Eng.Sun.COM: 970 971 >> Prior to SunOS 5.4, any LD_LIBRARY_PATH setting was ignored by the 972 >> runtime linker if the application was setxid (secure), thus your 973 >> applications search path would be: 974 >> 975 >> /usr/local/lib LD_LIBRARY_PATH component - IGNORED 976 >> /usr/lib LD_LIBRARY_PATH component - IGNORED 977 >> /usr/local/lib RPATH - honored 978 >> /usr/lib RPATH - honored 979 >> 980 >> the effect is that path 3 would be the first used, and this would 981 >> satisfy your resolv.so lookup. 982 >> 983 >> In SunOS 5.4 we made the LD_LIBRARY_PATH a little more flexible. 984 >> People who developed setxid applications wanted to be able to alter 985 >> the library search path to some degree to allow for their own 986 >> testing and debugging mechanisms. It was decided that the only 987 >> secure way to do this was to allow a `trusted' path to be used in 988 >> LD_LIBRARY_PATH. The only trusted directory we presently define 989 >> is /usr/lib. Thus a set-user-ID root developer could play with some 990 >> alternative shared object implementations and place them in 991 >> /usr/lib (being root we assume they'ed have access to write in this 992 >> directory). This change was made as part of 1155380 - after a 993 >> *huge* amount of discussion regarding the security aspect of things. 994 >> 995 >> So, in SunOS 5.4 your applications search path would be: 996 >> 997 >> /usr/local/lib from LD_LIBRARY_PATH - IGNORED (untrustworthy) 998 >> /usr/lib from LD_LIBRARY_PATH - honored (trustworthy) 999 >> /usr/local/lib from RPATH - honored 1000 >> /usr/lib from RPATH - honored 1001 >> 1002 >> here, path 2 would be the first used. 1003 1004Solaris 2.5.1 (SunOS 5.5.1) and 2.6 (SunOS 5.6) 1005 Apparently Solaris 2.5.1 patch 103663-01 installs a new 1006 /usr/include/resolv.h file that defines the __P macro without 1007 checking to see if it is already defined. This new resolv.h is also 1008 included in the Solaris 2.6 distribution. This causes compile 1009 warnings such as: 1010 1011 In file included from daemon.c:51: 1012 /usr/include/resolv.h:208: warning: `__P' redefined 1013 cdefs.h:58: warning: this is the location of the previous definition 1014 1015 These warnings can be safely ignored or you can create a resolv.h 1016 file in the obj.SunOS.5.5.1.* or obj.SunOS.5.6.* directory that reads: 1017 1018 #undef __P 1019 #include "/usr/include/resolv.h" 1020 1021 This problem was fixed in Solaris 7 (Sun bug ID 4081053). 1022 1023Solaris 7 (SunOS 5.7) 1024 Solaris 7 includes LDAP libraries but the implementation was 1025 lacking a few things. The following settings can be placed in 1026 devtools/Site/site.SunOS.5.7.m4 if you plan on using those 1027 libraries. 1028 1029 APPENDDEF(`confMAPDEF', `-DLDAPMAP') 1030 APPENDDEF(`confENVDEF', `-DLDAP_VERSION_MAX=3') 1031 APPENDDEF(`confLIBS', `-lldap') 1032 1033 Also, Sun's patch 107555 is needed to prevent a crash in the call 1034 to ldap_set_option for LDAP_OPT_REFERRALS in ldapmap_setopts if 1035 LDAP support is compiled in sendmail. 1036 1037Solaris 8 and later (SunOS 5.8 and later) 1038 Solaris 8 and later can optionally install LDAP support. If you 1039 have installed the Entire Distribution meta-cluster, you can use 1040 the following in devtools/Site/site.SunOS.5.8.m4 (or other 1041 appropriately versioned file) to enable LDAP: 1042 1043 APPENDDEF(`confMAPDEF', `-DLDAPMAP') 1044 APPENDDEF(`confLIBS', `-lldap') 1045 1046Solaris 9 and later (SunOS 5.9 and later) 1047 Solaris 9 and later have a revised LDAP library, libldap.so.5, 1048 which is derived from a Netscape implementation, thus requiring 1049 that SM_CONF_LDAP_MEMFREE be defined in conjunction with LDAPMAP: 1050 1051 APPENDDEF(`confMAPDEF', `-DLDAPMAP') 1052 APPENDDEF(`confENVDEF', `-DSM_CONF_LDAP_MEMFREE') 1053 APPENDDEF(`confLIBS', `-lldap') 1054 1055Solaris 1056 If you are using dns for hostname resolution on Solaris, make sure 1057 that the 'dns' entry is last on the hosts line in 1058 '/etc/nsswitch.conf'. For example, use: 1059 1060 hosts: nisplus files dns 1061 1062 Do not use: 1063 1064 hosts: nisplus dns [NOTFOUND=return] files 1065 1066 Note that 'nisplus' above is an illustration. The same comment 1067 applies no matter what naming services you are using. If you have 1068 anything other than dns last, even after "[NOTFOUND=return]", 1069 sendmail may not be able to determine whether an error was 1070 temporary or permanent. The error returned by the solaris 1071 gethostbyname() is the error for the last lookup used, and other 1072 naming services do not have the same concept of temporary failure. 1073 1074Ultrix 1075 By default, the IDENT protocol is turned off on Ultrix. If you 1076 are running Ultrix 4.4 or later, or if you have included patch 1077 CXO-8919 for Ultrix 4.2 or 4.3 to fix the TCP problem, you can turn 1078 IDENT on in the configuration file by setting the "ident" timeout. 1079 1080 The Ultrix 4.5 Y2K patch (ULTV45-022-1) has changed the resolver 1081 included in libc.a. Unfortunately, the __RES symbol hasn't changed 1082 and therefore, sendmail can no longer automatically detect the 1083 newer version. If you get a compiler error: 1084 1085 /lib/libc.a(gethostent.o): local_hostname_length: multiply defined 1086 1087 Then rebuild with this in devtools/Site/site.ULTRIX.m4: 1088 1089 APPENDDEF(`conf_sendmail_ENVDEF', `-DNEEDLOCAL_HOSTNAME_LENGTH=0') 1090 1091Digital UNIX (formerly DEC OSF/1) 1092 If you are compiling on OSF/1 (DEC Alpha), you must use 1093 -L/usr/shlib (otherwise it core dumps on startup). You may also 1094 need -mld to get the nlist() function, although some versions 1095 apparently don't need this. 1096 1097 Also, the enclosed makefile removed /usr/sbin/smtpd; if you need 1098 it, just create the link to the sendmail binary. 1099 1100 On DEC OSF/1 3.2 or earlier, the MatchGECOS option doesn't work 1101 properly due to a bug in the getpw* routines. If you want to use 1102 this, use -DDEC_OSF_BROKEN_GETPWENT=1. The problem is fixed in 3.2C. 1103 1104 Digital's mail delivery agent, /bin/mail (aka /bin/binmail), will 1105 only preserve the envelope sender in the "From " header if 1106 DefaultUserID is set to daemon. Setting this to mailnull will 1107 cause all mail to have the header "From mailnull ...". To use 1108 a different DefaultUserID, you will need to use a different mail 1109 delivery agent (such as mail.local found in the sendmail 1110 distribution). 1111 1112 On Digital UNIX 4.0 and later, Berkeley DB 1.85 is included with the 1113 operating system and already has the ndbm.o module removed. However, 1114 Digital has modified the original Berkeley DB db.h include file. 1115 This results in the following warning while compiling map.c and udb.c: 1116 1117 cc: Warning: /usr/include/db.h, line 74: The redefinition of the macro 1118 "__signed" conflicts with a current definition because the replacement 1119 lists differ. The redefinition is now in effect. 1120 #define __signed signed 1121 ------------------------^ 1122 1123 This warning can be ignored. 1124 1125 Digital UNIX's linker checks /usr/ccs/lib/ before /usr/lib/. 1126 If you have installed a new version of BIND in /usr/include 1127 and /usr/lib, you will experience difficulties as Digital ships 1128 libresolv.a in /usr/ccs/lib/ as well. Be sure to replace both 1129 copies of libresolv.a. 1130 1131IRIX 1132 The header files on SGI IRIX are completely prototyped, and as 1133 a result you can sometimes get some warning messages during 1134 compilation. These can be ignored. There are two errors in 1135 deliver only if you are using gcc, both of the form ``warning: 1136 passing arg N of `execve' from incompatible pointer type''. 1137 Also, if you compile with -DNIS, you will get a complaint 1138 about a declaration of struct dom_binding in a prototype 1139 when compiling map.c; this is not important because the 1140 function being prototyped is not used in that file. 1141 1142 In order to compile sendmail you will have had to install 1143 the developers' option in order to get the necessary include 1144 files. 1145 1146 If you compile with -lmalloc (the fast memory allocator), you may 1147 get warning messages such as the following: 1148 1149 ld32: WARNING 85: definition of _calloc in /usr/lib32/libmalloc.so 1150 preempts that definition in /usr/lib32/mips3/libc.so. 1151 ld32: WARNING 85: definition of _malloc in /usr/lib32/libmalloc.so 1152 preempts that definition in /usr/lib32/mips3/libc.so. 1153 ld32: WARNING 85: definition of _realloc in /usr/lib32/libmalloc.so 1154 preempts that definition in /usr/lib32/mips3/libc.so. 1155 ld32: WARNING 85: definition of _free in /usr/lib32/libmalloc.so 1156 preempts that definition in /usr/lib32/mips3/libc.so. 1157 ld32: WARNING 85: definition of _cfree in /usr/lib32/libmalloc.so 1158 preempts that definition in /usr/lib32/mips3/libc.so. 1159 1160 These are unavoidable and innocuous -- just ignore them. 1161 1162 According to Dave Sill <de5@ornl.gov>, there is a version of the 1163 Berkeley DB library patched to run on Irix 6.2 available from 1164 http://reality.sgi.com/ariel/freeware/#db . 1165 1166IRIX 6.x 1167 If you are using XFS filesystem, avoid using the -32 ABI switch to 1168 the cc compiler if possible. 1169 1170 Broken inet_aton and inet_ntoa on IRIX using gcc: There's 1171 a problem with gcc on IRIX, i.e., gcc can't pass structs 1172 less than 16 bits long unless they are 8 bits; IRIX 6.2 has 1173 some other sized structs. See 1174 http://www.bitmechanic.com/mail-archives/mysql/current/0418.html 1175 This problem seems to be fixed by gcc v2.95.2, gcc v2.8.1 1176 is reported as broken. Check your gcc version for this bug 1177 before installing sendmail. 1178 1179IRIX 6.4 1180 The IRIX 6.5.4 version of /bin/m4 does not work properly with 1181 sendmail. Either install fw_m4.sw.m4 off the Freeware_May99 CD and 1182 use /usr/freeware/bin/m4 or install and use GNU m4. 1183 1184NeXT or NEXTSTEP 1185 NEXTSTEP 3.3 and earlier ship with the old DBM library. Also, 1186 Berkeley DB does not currently run on NEXTSTEP. 1187 1188 If you are compiling on NEXTSTEP, you will have to create an 1189 empty file "unistd.h" and create a file "dirent.h" containing: 1190 1191 #include <sys/dir.h> 1192 #define dirent direct 1193 1194 (devtools/OS/NeXT should try to do both of these for you.) 1195 1196 Apparently, there is a bug in getservbyname on Nextstep 3.0 1197 that causes it to fail under some circumstances with the 1198 message "SYSERR: service "smtp" unknown" logged. You should 1199 be able to work around this by including the line: 1200 1201 OOPort=25 1202 1203 in your .cf file. 1204 1205BSDI (BSD/386) 1.0, NetBSD 0.9, FreeBSD 1.0 1206 The "m4" from BSDI won't handle the config files properly. 1207 I haven't had a chance to test this myself. 1208 1209 The M4 shipped in FreeBSD and NetBSD 0.9 don't handle the config 1210 files properly. One must use either GNU m4 1.1 or the PD-M4 1211 recently posted in comp.os.386bsd.bugs (and maybe others). 1212 NetBSD-current includes the PD-M4 (as stated in the NetBSD file 1213 CHANGES). 1214 1215 FreeBSD 1.0 RELEASE has uname(2) now. Use -DUSEUNAME in order to 1216 use it (look into devtools/OS/FreeBSD). NetBSD-current may have 1217 it too but it has not been verified. 1218 1219 The latest version of Berkeley DB uses a different naming 1220 scheme than the version that is supplied with your release. This 1221 means you will be able to use the current version of Berkeley DB 1222 with sendmail as long you use the new db.h when compiling 1223 sendmail and link it against the new libdb.a or libdb.so. You 1224 should probably keep the original db.h in /usr/include and the 1225 new db.h in /usr/local/include. 1226 12274.3BSD 1228 If you are running a "virgin" version of 4.3BSD, you'll have 1229 a very old resolver and be missing some header files. The 1230 header files are simple -- create empty versions and everything 1231 will work fine. For the resolver you should really port a new 1232 version (4.8.3 or later) of the resolver; 4.9 is available on 1233 gatekeeper.DEC.COM in pub/BSD/bind/4.9. If you are really 1234 determined to continue to use your old, buggy version (or as 1235 a shortcut to get sendmail working -- I'm sure you have the 1236 best intentions to port a modern version of BIND), you can 1237 copy ../contrib/oldbind.compat.c into sendmail and add the 1238 following to devtools/Site/site.config.m4: 1239 1240 APPENDDEF(`confOBJADD', `oldbind.compat.o') 1241 1242OpenBSD (up to 2.9 Release), NetBSD, FreeBSD (up to 4.3-RELEASE) 1243 m4 from *BSD won't handle libsm/Makefile.m4 properly, since the 1244 maximum length for strings is too short. You need to use GNU m4 1245 or patch m4, see for example: 1246 http://FreeBSD.org/cgi/cvsweb.cgi/src/usr.bin/m4/eval.c.diff?r1=1.11&r2=1.12 1247 1248A/UX 1249 Date: Tue, 12 Oct 1993 18:28:28 -0400 (EDT) 1250 From: "Eric C. Hagberg" <hagberg@med.cornell.edu> 1251 Subject: Fix for A/UX ndbm 1252 1253 I guess this isn't really a sendmail bug, however, it is something 1254 that A/UX users should be aware of when compiling sendmail 8.6. 1255 1256 Apparently, the calls that sendmail is using to the ndbm routines 1257 in A/UX 3.0.x contain calls to "broken" routines, in that the 1258 aliases database will break when it gets "just a little big" 1259 (sorry I don't have exact numbers here, but it broke somewhere 1260 around 20-25 aliases for me.), making all aliases non-functional 1261 after exceeding this point. 1262 1263 What I did was to get the gnu-dbm-1.6 package, compile it, and 1264 then re-compile sendmail with "-lgdbm", "-DNDBM", and using the 1265 ndbm.h header file that comes with the gnu-package. This makes 1266 things behave properly. 1267 [NOTE: see comment above about GDBM] 1268 1269 I suppose porting the New Berkeley DB package is another route, 1270 however, I made a quick attempt at it, and found it difficult 1271 (not easy at least); the gnu-dbm package "configured" and 1272 compiled easily. 1273 1274 [NOTE: Berkeley DB version 2.X runs on A/UX and can be used for 1275 database maps.] 1276 1277SCO Unix 1278 From: Thomas Essebier <tom@stallion.oz.au> 1279 Organisation: Stallion Technologies Pty Ltd. 1280 1281 It will probably help those who are trying to configure sendmail 8.6.9 1282 to know that if they are on SCO, they had better set 1283 OI-dnsrch 1284 or they will core dump as soon as they try to use the resolver. 1285 i.e., although SCO has _res.dnsrch defined, and is kinda BIND 4.8.3, 1286 it does not inititialise it, nor does it understand 'search' in 1287 /etc/named.boot. 1288 - sigh - 1289 1290 According to SCO, the m4 which ships with UnixWare 2.1.2 is broken. 1291 We recommend installing GNU m4 before attempting to build sendmail. 1292 1293 On some versions a bogus error value is listed if connections 1294 time out (large negative number). To avoid this explicitly set 1295 Timeout.connect to a reasonable value (several minutes). 1296 1297DG/UX 1298 Doug Anderson <dlander@afterlife.ncsc.mil> has successfully run 1299 V8 on the DG/UX 5.4.2 and 5.4R3.x platforms under heavy usage. 1300 Originally, the DG /bin/mail program wasn't compatible with 1301 the V8 sendmail, since the DG /bin/mail requires the environment 1302 variable "_FORCE_MAIL_LOCAL_=yes" be set. Version 8.7 now includes 1303 this in the environment before invoking the local mailer. Some 1304 have used procmail to avoid this problem in the past. It works 1305 but some have experienced file locking problems with their DG/UX 1306 ports of procmail. 1307 1308Apollo DomainOS 1309 If you are compiling on Apollo, you will have to create an empty 1310 file "unistd.h" (for DomainOS 10.3 and earlier) and create a file 1311 "dirent.h" containing: 1312 1313 #include <sys/dir.h> 1314 #define dirent direct 1315 1316 (devtools/OS/DomainOS will attempt to do both of these for you.) 1317 1318HP-UX 8.00 1319 Date: Mon, 24 Jan 1994 13:25:45 +0200 1320 From: Kimmo Suominen <Kimmo.Suominen@lut.fi> 1321 Subject: 8.6.5 w/ HP-UX 8.00 on s300 1322 1323 Just compiled and fought with sendmail 8.6.5 on a HP9000/360 (i.e., 1324 a series 300 machine) running HP-UX 8.00. 1325 1326 I was getting segmentation fault when delivering to a local user. 1327 With debugging I saw it was faulting when doing _free@libc... *sigh* 1328 It seems the new implementation of malloc on s300 is buggy as of 8.0, 1329 so I tried out the one in -lmalloc (malloc(3X)). With that it seems 1330 to work just dandy. 1331 1332 When linking, you will get the following error: 1333 1334 ld: multiply defined symbol _freespace in file /usr/lib/libmalloc.a 1335 1336 but you can just ignore it. You might want to add this info to the 1337 README file for the future... 1338 1339Linux 1340 Something broke between versions 0.99.13 and 0.99.14 of Linux: the 1341 flock() system call gives errors. If you are running .14, you must 1342 not use flock. You can do this with -DHASFLOCK=0. We have also 1343 been getting complaints since version 2.4.X was released. 1344 sendmail 8.13 has changed the default locking method to fcntl() 1345 for Linux kernel version 2.4 and later. Be sure to update other 1346 sendmail related programs to match locking techniques (some 1347 examples, besides makemap and mail.local, include procmail, mailx, 1348 mutt, elm, etc). 1349 1350 Around the inclusion of bind-4.9.3 & Linux libc-4.6.20, the 1351 initialization of the _res structure changed. If /etc/hosts.conf 1352 was configured as "hosts, bind" the resolver code could return 1353 "Name server failure" errors. This is supposedly fixed in 1354 later versions of libc (>= 4.6.29?), and later versions of 1355 sendmail (> 8.6.10) try to work around the problem. 1356 1357 Some older versions (< 4.6.20?) of the libc/include files conflict 1358 with sendmail's version of cdefs.h. Deleting sendmail's version 1359 on those systems should be non-harmful, and new versions don't care. 1360 1361 NOTE ON LINUX & BIND: By default, the Makefile generated for Linux 1362 includes header files in /usr/local/include and libraries in 1363 /usr/local/lib. If you've installed BIND on your system, the header 1364 files typically end up in the search path and you need to add 1365 "-lresolv" to the LIBS line in your Makefile. Really old versions 1366 may need to include "-l44bsd" as well (particularly if the link phase 1367 complains about missing strcasecmp, strncasecmp or strpbrk). 1368 Complaints about an undefined reference to `__dn_skipname' in 1369 domain.o are a sure sign that you need to add -lresolv to LIBS. 1370 Newer versions of Linux are basically threaded BIND, so you may or 1371 may not see complaints if you accidentally mix BIND 1372 headers/libraries with virginal libc. If you have BIND headers in 1373 /usr/local/include (resolv.h, etc) you *should* be adding -lresolv 1374 to LIBS. Data structures may change and you'd be asking for a 1375 core dump. 1376 1377 A number of problems have been reported regarding the Linux 2.2.0 1378 kernel. So far, these problems have been tracked down to syslog() 1379 and DNS resolution. We believe the problem is with the poll() 1380 implementation in the Linux 2.2.0 kernel and poll()-aware versions 1381 of glib (at least up to 2.0.111). 1382 1383glibc 1384 glibc 2.2.1 (and possibly other versions) changed the value of 1385 __RES in resolv.h but failed to actually provide the IPv6 API 1386 changes that the change implied. Therefore, compiling with 1387 -DNETINET6 fails. 1388 1389 Workarounds: 1390 1) Compile without -DNETINET6 1391 2) Build against a real BIND 8.2.2 include/lib tree 1392 3) Wait for glibc to fix it 1393 1394AIX 4.X 1395 The AIX 4.X linker uses library paths specified during compilation 1396 using -L for run-time shared library searches. Therefore, it is 1397 vital that relative and unsafe directory paths not be using when 1398 compiling sendmail. Because of this danger, by default, compiles 1399 on AIX use the -blibpath option to limit shared libraries to 1400 /usr/lib and /lib. If you need to allow more directories, such as 1401 /usr/local/lib, modify your devtools/Site/site.AIX.4.2.m4, 1402 site.AIX.4.3.m4, and/or site.AIX.4.x.m4 file(s) and set confLDOPTS 1403 appropriately. For example: 1404 1405 define(`confLDOPTS', `-blibpath:/usr/lib:/lib:/usr/local/lib') 1406 1407 Be sure to only add (safe) system directories. 1408 1409 The AIX version of GNU ld also exhibits this problem. If you are 1410 using that version, instead of -blibpath, use its -rpath option. 1411 For example: 1412 1413 gcc -Wl,-rpath /usr/lib -Wl,-rpath /lib -Wl,-rpath /usr/local/lib 1414 1415AIX 4.X If the test program t-event (and most others) in libsm fails, 1416 check your compiler settings. It seems that the flags -qnoro or 1417 -qnoroconst on some AIX versions trigger a compiler bug. Check 1418 your compiler settings or use cc instead of xlc. 1419 1420AIX 4.0-4.2, maybe some AIX 4.3 versions 1421 The AIX m4 implements a different mechanism for ifdef which is 1422 inconsistent with other versions of m4. Therefore, it will not 1423 work properly with the sendmail Build architecture or m4 1424 configuration method. To work around this problem, please use 1425 GNU m4 from ftp://ftp.gnu.org/pub/gnu/. 1426 The problem seems to be solved in AIX 4.3.3 at least. 1427 1428AIX 4.3.3 1429 From: Valdis.Kletnieks@vt.edu 1430 Date: Sun, 02 Jul 2000 03:58:02 -0400 1431 1432 Under AIX 4.3.3, after applying bos.adt.include 4.3.3.12 to close the 1433 BIND 8.2.2 security holes, you can no longer build with -DNETINET6 1434 because they changed the value of __RES in resolv.h but failed to 1435 actually provide the API changes that the change implied. 1436 1437 Workarounds: 1438 1) Compile without -DNETINET6 1439 2) Build against a real BIND 8.2.2 include/lib tree 1440 3) Wait for IBM to fix it 1441 1442AIX 3.x 1443 This version of sendmail does not support MB, MG, and MR resource 1444 records, which are supported by AIX sendmail. 1445 1446 Several people have reported that the IBM-supplied named returns 1447 fairly random results -- the named should be replaced. It is not 1448 necessary to replace the resolver, which will simplify installation. 1449 A new BIND resolver can be found at http://www.isc.org/isc/. 1450 1451AIX 3.1.x 1452 The supplied load average code only works correctly for AIX 3.2.x. 1453 For 3.1, use -DLA_TYPE=LA_SUBR and get the latest ``monitor'' 1454 package by Jussi Maki <jmaki@hut.fi> from ftp.funet.fi in the 1455 directory pub/unix/AIX/rs6000/monitor-1.12.tar.Z; use the loadavgd 1456 daemon, and the getloadavg subroutine supplied with that package. 1457 If you don't care about load average throttling, just turn off 1458 load average checking using -DLA_TYPE=LA_ZERO. 1459 1460RISC/os 1461 RISC/os from MIPS is a merged AT&T/Berkeley system. When you 1462 compile on that platform you will get duplicate definitions 1463 on many files. You can ignore these. 1464 1465System V Release 4 Based Systems 1466 There is a single devtools OS that is intended for all SVR4-based 1467 systems (built from devtools/OS/SVR4). It defines __svr4__, 1468 which is predefined by some compilers. If your compiler already 1469 defines this compile variable, you can delete the definition from 1470 the generated Makefile or create a devtools/Site/site.config.m4 1471 file. 1472 1473 It's been tested on Dell Issue 2.2. 1474 1475DELL SVR4 1476 Date: Mon, 06 Dec 1993 10:42:29 EST 1477 From: "Kimmo Suominen" <kim@grendel.lut.fi> 1478 Message-ID: <2d0352f9.lento29@lento29.UUCP> 1479 To: eric@cs.berkeley.edu 1480 Cc: sendmail@cs.berkeley.edu 1481 Subject: Notes for DELL SVR4 1482 1483 Eric, 1484 1485 Here are some notes for compiling Sendmail 8.6.4 on DELL SVR4. I ran 1486 across these things when helping out some people who contacted me by 1487 e-mail. 1488 1489 1) Use gcc 2.4.5 (or later?). Dell distributes gcc 2.1 with their 1490 Issue 2.2 Unix. It is too old, and gives you problems with 1491 clock.c, because sigset_t won't get defined in <sys/signal.h>. 1492 This is due to a problematic protection rule in there, and is 1493 fixed with gcc 2.4.5. 1494 1495 2) If you don't use the new Berkeley DB (-DNEWDB), then you need 1496 to add "-lc -lucb" to the libraries to link with. This is because 1497 the -ldbm distributed by Dell needs the bcopy, bcmp and bzero 1498 functions. It is important that you specify both libraries in 1499 the given order to be sure you only get the BSTRING functions 1500 from the UCB library (and not the signal routines etc.). 1501 1502 3) Don't leave out "-lelf" even if compiling with "-lc -lucb". 1503 The UCB library also has another copy of the nlist routines, 1504 but we do want the ones from "-lelf". 1505 1506 If anyone needs a compiled gcc 2.4.5 and/or a ported DB library, they 1507 can use anonymous ftp to fetch them from lut.fi in the /kim directory. 1508 They are copies of what I use on grendel.lut.fi, and offering them 1509 does not imply that I would also support them. I have sent the DB 1510 port for SVR4 back to Keith Bostic for inclusion in the official 1511 distribution, but I haven't heard anything from him as of today. 1512 1513 - gcc-2.4.5-svr4.tar.gz (gcc 2.4.5 and the corresponding libg++) 1514 - db-1.72.tar.gz (with source, objects and a installed copy) 1515 1516 Cheers 1517 + Kim 1518 -- 1519 * Kimmo.Suominen@lut.fi * SysVr4 enthusiast at GRENDEL.LUT.FI * 1520 * KIM@FINFILES.BITNET * Postmaster and Hostmaster at LUT.FI * 1521 * + 358 200 865 718 * Unix area moderator at NIC.FUNET.FI * 1522 1523ConvexOS 10.1 and below 1524 In order to use the name server, you must create the file 1525 /etc/use_nameserver. If this file does not exist, the call 1526 to res_init() will fail and you will have absolutely no 1527 access to DNS, including MX records. 1528 1529Amdahl UTS 2.1.5 1530 In order to get UTS to work, you will have to port BIND 4.9. 1531 The vendor's BIND is reported to be ``totally inadequate.'' 1532 See sendmail/contrib/AmdahlUTS.patch for the patches necessary 1533 to get BIND 4.9 compiled for UTS. 1534 1535UnixWare 1536 According to Alexander Kolbasov <sasha@unitech.gamma.ru>, 1537 the m4 on UnixWare 2.0 (still in Beta) will core dump on the 1538 config files. GNU m4 and the m4 from UnixWare 1.x both work. 1539 1540 According to Larry Rosenman <ler@lerami.lerctr.org>: 1541 1542 UnixWare 2.1.[23]'s m4 chokes (not obviously) when 1543 processing the 8.9.0 cf files. 1544 1545 I had a LOCAL_RULE_0 that wound up AFTER the 1546 SBasic_check_rcpt rules using the SCO supplied M4. 1547 GNU M4 works fine. 1548 1549UNICOS 8.0.3.4 1550 Some people have reported that the -O flag on UNICOS can cause 1551 problems. You may want to turn this off if you have problems 1552 running sendmail. Reported by Jerry G. DeLapp <jgd@acl.lanl.gov>. 1553 1554Darwin/Mac OS X (10.X.X) 1555 The linker errors produced regarding getopt() and its associated 1556 variables can safely be ignored. 1557 1558 From Mike Zimmerman <zimmy@torrentnet.com>: 1559 1560 From scratch here is what Darwin users need to do to the standard 1561 10.0.0, 10.0.1 install to get sendmail working. 1562 From http://www.macosx.com/forums/showthread.php?s=6dac0e9e1f3fd118a4870a8a9b559491&threadid=2242: 1563 1. chmod g-w / /private /private/etc 1564 2. Properly set HOSTNAME in /etc/hostconfig to your FQDN: 1565 HOSTNAME=-my.domain.com- 1566 3. Edit /etc/rc.boot: 1567 hostname my.domain.com 1568 domainname domain.com 1569 4. Edit /System/Library/StartupItems/Sendmail/Sendmail: 1570 Remove the "&" after the sendmail command: 1571 /usr/sbin/sendmail -bd -q1h 1572 1573 From Carsten Klapp <carsten.klapp@home.com>: 1574 1575 The easiest workaround is to remove the group-writable permission 1576 for the root directory and the symbolic /etc inherits this 1577 change. While this does fix sendmail, the unfortunate side-effect 1578 is the OS X admin will no longer be able to manipulate icons in the 1579 top level of the Startup disk unless logged into the GUI as the 1580 superuser. 1581 1582 In applying the alternate workaround, care must be taken while 1583 swapping the symlink /etc with the directory /private/etc. In all 1584 likelihood any admin who is concerned with this sendmail error has 1585 enough experience to not accidentally harm anything in the process. 1586 1587 a. Swap the /etc symlink with /private/etc (as superuser): 1588 rm /etc 1589 mv /private/etc /etc 1590 ln -s /etc /private/etc 1591 1592 b. Set / to group unwritable (as superuser): 1593 chmod g-w / 1594 1595Darwin/Mac OS X (10.1.5) 1596 Apple's upgrade to sendmail 8.12 is incorrectly configured. You 1597 will need to manually fix it up by doing the following: 1598 1599 1. chown smmsp:smmsp /var/spool/clientmqueue 1600 2. chmod 2770 /var/spool/clientmqueue 1601 3. chgrp smmsp /usr/sbin/sendmail 1602 4. chmod g+s /usr/sbin/sendmail 1603 1604 From Daniel J. Luke <dluke@geeklair.net>: 1605 1606 It appears that setting the sendmail.cf property in 1607 /locations/sendmail in NetInfo on Mac OS X 10.1.5 with sendmail 1608 8.12.4 causes 'bad things' to happen. 1609 1610 Specifically sendmail instances that should be getting their config 1611 from /etc/mail/submit.cf don't (so mail/mutt/perl scripts which 1612 open pipes to sendmail stop working as sendmail tries to write to 1613 /var/spool/mqueue and cannot as sendmail is no longer suid root). 1614 1615 Removing the entry from NetInfo fixes this problem. 1616 1617GNU getopt 1618 I'm told that GNU getopt has a problem in that it gets confused 1619 by the double call. Use the version in conf.c instead. 1620 1621BIND 4.9.2 and Ultrix 1622 If you are running on Ultrix, be sure you read conf/Info.Ultrix 1623 in the BIND distribution very carefully -- there is information 1624 in there that you need to know in order to avoid errors of the 1625 form: 1626 1627 /lib/libc.a(gethostent.o): sethostent: multiply defined 1628 /lib/libc.a(gethostent.o): endhostent: multiply defined 1629 /lib/libc.a(gethostent.o): gethostbyname: multiply defined 1630 /lib/libc.a(gethostent.o): gethostbyaddr: multiply defined 1631 1632 during the link stage. 1633 1634BIND 8.X 1635 BIND 8.X returns HOST_NOT_FOUND instead of TRY_AGAIN on temporary 1636 DNS failures when trying to find the hostname associated with an IP 1637 address (gethostbyaddr()). This can cause problems as 1638 $&{client_name} based lookups in class R ($=R) and the access 1639 database won't succeed. 1640 1641 This will be fixed in BIND 8.2.1. For earlier versions, this can 1642 be fixed by making "dns" the last name service queried for host 1643 resolution in /etc/irs.conf: 1644 1645 hosts local continue 1646 hosts dns 1647 1648strtoul 1649 Some compilers (notably gcc) claim to be ANSI C but do not 1650 include the ANSI-required routine "strtoul". If your compiler 1651 has this problem, you will get an error in srvrsmtp.c on the 1652 code: 1653 1654 # ifdef defined(__STDC__) && !defined(BROKEN_ANSI_LIBRARY) 1655 e->e_msgsize = strtoul(vp, (char **) NULL, 10); 1656 # else 1657 e->e_msgsize = strtol(vp, (char **) NULL, 10); 1658 # endif 1659 1660 You can use -DBROKEN_ANSI_LIBRARY to get around this problem. 1661 1662Listproc 6.0c 1663 Date: 23 Sep 1995 23:56:07 GMT 1664 Message-ID: <95925101334.~INN-AUMa00187.comp-news@dl.ac.uk> 1665 From: alansz@mellers1.psych.berkeley.edu (Alan Schwartz) 1666 Subject: Listproc 6.0c + Sendmail 8.7 [Helpful hint] 1667 1668 Just upgraded to sendmail 8.7, and discovered that listproc 6.0c 1669 breaks, because it, by default, sends a blank "HELO" rather than 1670 a "HELO hostname" when using the 'system' or 'telnet' mail method. 1671 1672 The fix is to include -DZMAILER in the compilation, which will 1673 cause it to use "HELO hostname" (which Z-mail apparently requires 1674 as well. :) 1675 1676OpenSSL 1677 OpenSSL versions prior to 0.9.6 use a macro named Free which 1678 conflicts with existing macro names on some platforms, such as 1679 AIX. 1680 Do not use 0.9.3, but OpenSSL 0.9.5a or later if compatible with 1681 0.9.5a. 1682 1683PH 1684 PH support is provided by Mark Roth <roth@uiuc.edu>. The map is 1685 described at http://www-dev.cites.uiuc.edu/sendmail/ . 1686 1687 NOTE: The "spacedname" pseudo-field which was used by earlier 1688 versions of the PH map code is no longer supported! See the URL 1689 listed above for more information. 1690 1691 Please contact Mark Roth for support and questions regarding the 1692 map. 1693 1694TCP Wrappers 1695 If you are using -DTCPWRAPPERS to get TCP Wrappers support you will 1696 also need to install libwrap.a and modify your site.config.m4 file 1697 or the generated Makefile to include -lwrap in the LIBS line 1698 (make sure that INCDIRS and LIBDIRS point to where the tcpd.h and 1699 libwrap.a can be found). 1700 1701 TCP Wrappers is available at ftp://ftp.porcupine.org/pub/security/. 1702 1703 If you have alternate MX sites for your site, be sure that all of 1704 your MX sites reject the same set of hosts. If not, a bad guy whom 1705 you reject will connect to your site, fail, and move on to the next 1706 MX site, which will accept the mail for you and forward it on to you. 1707 1708Regular Expressions (MAP_REGEX) 1709 If sendmail linking fails with: 1710 1711 undefined reference to 'regcomp' 1712 1713 or sendmail gives an error about a regular expression with: 1714 1715 pattern-compile-error: : Operation not applicable 1716 1717 Your libc does not include a running version of POSIX-regex. Use 1718 librx or regex.o from the GNU Free Software Foundation, 1719 ftp://ftp.gnu.org/pub/gnu/rx-?.?.tar.gz or 1720 ftp://ftp.gnu.org/pub/gnu/regex-?.?.tar.gz. 1721 You can also use the regex-lib by Henry Spencer, 1722 ftp://ftp.funet.fi/pub/languages/C/spencer/regex.shar.gz 1723 Make sure, your compiler reads regex.h from the distribution, 1724 not from /usr/include, otherwise sendmail will dump a core. 1725 1726Fedora Core 5, 64 bit version 1727 If the ld stage fails with undefined functions like 1728 __res_querydomain, __dn_expand 1729 then add these lines to devtools/Site/site.config.m4 1730 1731 APPENDDEF(`confLIBDIRS', `-L/usr/lib64') 1732 APPENDDEF(`confINCDIRS', `-I/usr/include/bind9') 1733 1734 and rebuild (sh ./Build -c). 1735 1736 Problem noted by Daniel Krones, solution suggested by 1737 Anthony Howe. 1738 1739 1740+--------------+ 1741| MANUAL PAGES | 1742+--------------+ 1743 1744The manual pages have been written against the -man macros, and 1745should format correctly with any reasonable *roff. 1746 1747 1748+-----------------+ 1749| DEBUGGING HOOKS | 1750+-----------------+ 1751 1752As of 8.6.5, sendmail daemons will catch a SIGUSR1 signal and log 1753some debugging output (logged at LOG_DEBUG severity). The 1754information dumped is: 1755 1756 * The value of the $j macro. 1757 * A warning if $j is not in the set $=w. 1758 * A list of the open file descriptors. 1759 * The contents of the connection cache. 1760 * If ruleset 89 is defined, it is evaluated and the results printed. 1761 1762This allows you to get information regarding the runtime state of the 1763daemon on the fly. This should not be done too frequently, since 1764the process of rewriting may lose memory which will not be recovered. 1765Also, ruleset 89 may call non-reentrant routines, so there is a small 1766non-zero probability that this will cause other problems. It is 1767really only for debugging serious problems. 1768 1769A typical formulation of ruleset 89 would be: 1770 1771 R$* $@ $>0 some test address 1772 1773 1774+-----------------------------+ 1775| DESCRIPTION OF SOURCE FILES | 1776+-----------------------------+ 1777 1778The following list describes the files in this directory: 1779 1780Build Shell script for building sendmail. 1781Makefile A convenience for calling ./Build. 1782Makefile.m4 A template for constructing a makefile based on the 1783 information in the devtools directory. 1784README This file. 1785TRACEFLAGS My own personal list of the trace flags -- not guaranteed 1786 to be particularly up to date. 1787alias.c Does name aliasing in all forms. 1788aliases.5 Man page describing the format of the aliases file. 1789arpadate.c A subroutine which creates ARPANET standard dates. 1790bf.c Routines to implement memory-buffered file system using 1791 hooks provided by libsm now (formerly Torek stdio library). 1792bf.h Buffered file I/O function declarations and 1793 data structure and function declarations for bf.c. 1794collect.c The routine that actually reads the mail into a temp 1795 file. It also does a certain amount of parsing of 1796 the header, etc. 1797conf.c The configuration file. This contains information 1798 that is presumed to be quite static and non- 1799 controversial, or code compiled in for efficiency 1800 reasons. Most of the configuration is in sendmail.cf. 1801conf.h Configuration that must be known everywhere. 1802control.c Routines to implement control socket. 1803convtime.c A routine to sanely process times. 1804daemon.c Routines to implement daemon mode. 1805deliver.c Routines to deliver mail. 1806domain.c Routines that interface with DNS (the Domain Name 1807 System). 1808envelope.c Routines to manipulate the envelope structure. 1809err.c Routines to print error messages. 1810headers.c Routines to process message headers. 1811helpfile An example helpfile for the SMTP HELP command and -bt mode. 1812macro.c The macro expander. This is used internally to 1813 insert information from the configuration file. 1814mailq.1 Man page for the mailq command. 1815main.c The main routine to sendmail. This file also 1816 contains some miscellaneous routines. 1817makesendmail A convenience for calling ./Build. 1818map.c Support for database maps. 1819mci.c Routines that handle mail connection information caching. 1820milter.c MTA portions of the mail filter API. 1821mime.c MIME conversion routines. 1822newaliases.1 Man page for the newaliases command. 1823parseaddr.c The routines which do address parsing. 1824queue.c Routines to implement message queueing. 1825readcf.c The routine that reads the configuration file and 1826 translates it to internal form. 1827recipient.c Routines that manipulate the recipient list. 1828sasl.c Routines to interact with Cyrys-SASL. 1829savemail.c Routines which save the letter on processing errors. 1830sendmail.8 Man page for the sendmail command. 1831sendmail.h Main header file for sendmail. 1832sfsasl.c I/O interface between SASL/TLS and the MTA. 1833sfsasl.h Header file for sfsasl.c. 1834shmticklib.c Routines for shared memory counters. 1835sm_resolve.c Routines for DNS lookups (for DNS map type). 1836sm_resolve.h Header file for sm_resolve.c. 1837srvrsmtp.c Routines to implement server SMTP. 1838stab.c Routines to manage the symbol table. 1839stats.c Routines to collect and post the statistics. 1840statusd_shm.h Data structure and function declarations for shmticklib.c. 1841sysexits.c List of error messages associated with error codes 1842 in sysexits.h. 1843sysexits.h List of error codes for systems that lack their own. 1844timers.c Routines to provide microtimers. 1845timers.h Data structure and function declarations for timers.h. 1846tls.c Routines for TLS. 1847trace.c The trace package. These routines allow setting and 1848 testing of trace flags with a high granularity. 1849udb.c The user database interface module. 1850usersmtp.c Routines to implement user SMTP. 1851util.c Some general purpose routines used by sendmail. 1852version.c The version number and information about this 1853 version of sendmail. 1854 1855(Version $Revision: 8.393 $, last update $Date: 2013-11-22 20:51:54 $ ) 1856