3 K N O W N B U G S I N S E N D M A I L
7 The following are bugs or deficiencies in sendmail that I am aware of
8 but which have not been fixed in the current release. You probably
9 want to get the most up to date version of this from ftp.sendmail.org
10 in /pub/sendmail/KNOWNBUGS. For descriptions of bugs that have been
11 fixed, see the file RELEASE_NOTES (in the root directory of the sendmail
14 This list is not guaranteed to be complete.
17 * Null bytes are not handled properly in headers.
19 Sendmail should handle full binary data. As it stands, it handles
20 all values in the body, but only 0x01-0x80 and 0xA0-0xFF in
21 the header. Notably missing is 0x00, which would require a major
22 restructuring of the code -- for example, almost no C library support
23 could be used to handle strings.
25 * Duplicate error messages.
27 Sometimes identical, duplicate error messages can be generated. As
28 near as I can tell, this is rare and relatively innocuous.
30 * $c (hop count) macro improperly set.
32 The $c macro is supposed to contain the current hop count, for use
33 when calling a mailer. This macro is initialized too early, and
34 is always zero (or the value of the -c command line flag, if any).
35 This macro will probably be removed entirely in a future release;
36 I don't believe there are any mailers left that require it.
38 * If you EXPN a list or user that has a program mailer, the output of
39 EXPN will include ``@local.host.name''. You can't actually mail to
40 this address. It's not clear what the right behavior is in this
43 * \231 considered harmful.
45 Header addresses that have the \231 character (and possibly others
46 in the range \201 - \237) behave in odd and usually unexpected ways.
48 * accept() problem on SVR4.
50 Apparently, the sendmail daemon loop (doing accept()s on the network)
51 can get into a weird state on SVR4; it starts logging ``SYSERR:
52 getrequests: accept: Protocol Error''. The workaround is to kill
53 and restart the sendmail daemon. We don't have an SVR4 system at
54 Berkeley that carries more than token mail load, so I can't validate
55 this. It is likely to be a glitch in the sockets emulation, since
56 "Protocol Error" is not possible error code with Berkeley TCP/IP.
58 I've also had someone report the message ``sendmail: accept:
59 SIOCGPGRP failed errno 22'' on an SVR4 system. This message is
60 not in the sendmail source code, so I assume it is also a bug
61 in the sockets emulation. (Errno 22 is EINVAL "Invalid Argument"
62 on all the systems I have available, including Solaris 2.x.)
63 Apparently, this problem is due to linking -lc before -lsocket;
64 if you are having this problem, check your Makefile.
66 * accept() problem on Linux.
68 The accept() in sendmail daemon loop can return ETIMEDOUT. An
69 error is reported to syslog:
71 Jun 9 17:14:12 hostname sendmail[207]: NOQUEUE: SYSERR(root):
72 getrequests: accept: Connection timed out
74 "Connection timed out" is not documented as a valid return from
75 accept(2) and this was believed to be a bug in the Linux kernel.
76 Later information from the Linux kernel group states that Linux
77 2.0 kernels follow RFC1122 while sendmail follows the original BSD
78 (now POSIX 1003.1g draft) specification. The 2.1.X and later kernels
79 will follow the POSIX draft.
81 * Excessive mailing list nesting can run out of file descriptors.
83 If you have a mailing list that includes lots of other mailing
84 lists, each of which has a separate owner, you can run out of
85 file descriptors. Each mailing list with a separate owner uses
86 one open file descriptor (prior to 8.6.6 it was three open
87 file descriptors per list). This is particularly egregious if
88 you have your connection cache set to be large.
90 * Connection caching breaks if you pass the port number as an argument.
92 If you have a definition such as:
94 Mport, P=[IPC], F=kmDFMuX, S=11/31, R=21,
95 M=2100000, T=DNS/RFC822/SMTP,
98 (i.e., where $h is the port number instead of the host name) the
99 connection caching code will break because it won't notice that
100 two messages addressed to different ports should use different
103 * ESMTP SIZE underestimates the size of a message
105 Sendmail makes no allowance for headers that it adds, nor does it
106 account for the SMTP on-the-wire \r\n expansion. It probably doesn't
107 allow for 8->7 bit MIME conversions either.
109 * Paths to programs being executed and the mode of program files are
110 not checked. Essentially, the RunProgramInUnsafeDirPath and
111 RunWritableProgram bits in the DontBlameSendmail option are always
112 set. This is not a problem if your system is well managed (that is,
113 if binaries and system directories are mode 755 instead of something
116 * 8-bit data in GECOS field
118 If the GECOS (personal name) information in the passwd file contains
119 8-bit characters, those characters can be included in the message
120 header, which can cause problems when sending SMTP to hosts that
121 only accept 7-bit characters.
123 * 8->7 bit MIME conversion
125 When sendmail is doing 8->7 bit MIME conversions, and the message
126 contains certain MIME body types that cannot be converted to 7-bit,
127 sendmail will strip the message to 7-bit.
129 * 7->8 bit MIME conversion
131 If a message that is encoded as 7-bit MIME is converted to 8-bit and
132 that message when decoded is illegal (e.g., because of long lines or
133 illegal characters), sendmail can produce an illegal message.
135 * MIME encoded full name phrases in the From: header
137 If a full name phrase includes characters from MustQuoteChars, sendmail
138 will quote the entire full name phrase. If MustQuoteChars includes
139 characters which are not special characters according to STD 11 (RFC
140 822), this quotation can interfere with MIME encoded full name phrases.
141 By default, sendmail includes the single quote character (') in
142 MustQuoteChars even though it is not listed as a special character in
145 * bestmx map with -z flag truncates the list of MX hosts
147 A bestmx map configured with the -z flag will truncate the list
148 of MX hosts. This prevents creation of strings which are too
149 long for ruleset parsing. This can have an adverse effect on the
150 relay_based_on_MX feature.
152 * Saving to ~sender/dead.letter fails if su'ed to root
154 If ErrorMode is set to print and an error in sending mail occurs,
155 the normal action is to print a message to the screen and append
156 the message to a dead.letter file in the sender's home directory.
157 In the case where the sender is using su to act as root, the file
158 safety checks prevent sendmail from saving the dead.letter file
159 because the sender's uid and the current real uid do not match.
161 * Berkeley DB 2.X race condition with fcntl() locking
163 There is a race condition for Berkeley DB 2.X databases on
164 operating systems which use fcntl() style locking, such as
165 Solaris. Sendmail locks the map before calling db_open() to
166 prevent others from modifying the map while it is being opened.
167 Unfortunately, Berkeley DB opens the map, closes it, and then
168 reopens it. fcntl() locking drops the lock when any file
169 descriptor pointing to the file is closed, even if it is a
170 different file descriptor than the one used to initially lock
171 the file. As a result there is a possibility that entries in a
172 map might not be found during a map rebuild. As a workaround,
173 you can use makemap to build a map with a new name and then
174 "mv" the new db file to replace the old one.
176 * File open timeouts not available on hard mounted NFS file systems
178 Since SIGALRM does not interrupt an RPC call for hard mounted
179 NFS file systems, it is impossible to implement a timeout on a file
180 open operation. Therefore, while the NFS server is not responding,
181 attempts to open a file on that server will hang. Systems with
182 local mail delivery and NFS hard mounted home directories should be
183 avoided, as attempts to open the forward files could hang.
185 (Version 8.36, last updated 2/4/1999)