xref: /csrg-svn/usr.sbin/sendmail/KNOWNBUGS (revision 67920)

	     K N O W N   B U G S   I N   S E N D M A I L
			     (for 8.6.9)


The following are bugs or deficiencies in sendmail that I am aware of
but which have not been fixed in the current release.  You probably
want to get the most up to date version of this from FTP.CS.Berkeley.EDU
in /ucb/sendmail/KNOWNBUGS.  For descriptions of bugs that have been
fixed, see the file RELEASE_NOTES (in the root directory of the sendmail
distribution).

This list is not guaranteed to be complete.


* Null bytes are not handled properly.

  Sendmail should handle full binary data.  As it stands, it handles
  any value from 0x01-0xFF in the body and 0x01-0x80 and 0xA0-0xFF in
  the header.  Notably missing is 0x00, which would require a major
  restructuring of the code -- for example, almost no C library support
  could be used to handle strings.  Fixed (for the message body only)
  in 8.7.

* Duplicate error messages.

  Sometimes identical, duplicate error messages can be generated.  As
  near as I can tell, this is rare and relatively innocuous.

* No "exposed users" in "nullrelay" configuration.

  The "nullrelay" configuration hides all addresses behind the mail
  hub name.  Some sites might prefer to expose some names such as
  root.  This information is always available in Received: lines.

* $c (hop count) macro improperly set.

  The $c macro is supposed to contain the current hop count, for use
  when calling a mailer.  This macro is initialized too early, and
  is always zero (or the value of the -c command line flag, if any).
  This macro will probably be removed entirely in a future release;
  I don't believe there are any mailers left that require it.

* If you EXPN a list or user that has a program mailer, the output of
  EXPN will include ``@local.host.name''.  You can't actually mail to
  this address.  It's not clear what the right behaviour is in this
  circumstance.

* REDIRECT aliases don't work with `n' option.

  If you have option `n' set when you use newaliases and have
  REDIRECT addresses in your aliases file, you'll get the error
  messages during the newaliases instead of when email is sent to
  the address in question.  The workaround is to turn off the `n'
  option.

* MX records that point at non-existent hosts work strangly.

  Consider the DNS records:

	hostH	MX	1 hostA
		MX	2 hostB
	hostA	A	128.32.8.9

  (note that there is no A record for hostB).  If hostA is down,
  an attempt to send to hostH gives "host unknown" -- that is, it
  reflects out the status on the last host it tries, which in this
  case is hostB, which is unknown.  It probably ought to eliminate
  hostB early in processing.

* NAME environment variables with commas and unbalanced quotes break.

  If you define your NAME environment variable to have a comma
  (e.g., ``Lastname, Firstname''), and you are using the $q definition
  that uses ``name <address>'' format, sendmail treats the first and
  last names as two addresses, thus producing a bogus From line.  You
  can work around this by changing the $q definition to use
  ``address (name)''.

  If you have an unbalanced double quote (e.g., ``Firstname"Lastname'')
  the auto-quoting algorithm breaks.

* \231 considered harmful.

  Header addresses that have the \231 character (and possibly others
  in the range \201 - \237) behave in odd and usually unexpected ways.

* DEC Alphas (OSF/1 1.3) sometimes time out on sending mail.

  I have one report that DEC Alphas acting as SMTP clients sometimes
  will apparently not see the "250 OK" message in response to the
  dot that indicates the end of the message.  This only happens if
  the message is run from the queue -- if it gets through on first
  try, everything is fine.  I have been unable to reproduce this
  problem at Berkeley.

* accept() problem on SVR4.

  Apparently, the sendmail daemon loop (doing accept()s on the network)
  can get into a wierd state on SVR4; it starts logging ``SYSERR:
  getrequests: accept: Protocol Error''.  The workaround is to kill
  and restart the sendmail daemon.  We don't have an SVR4 system at
  Berkeley that carries more than token mail load, so I can't validate
  this.  It is likely to be a glitch in the sockets emulation, since
  "Protocol Error" is not possible error code with Berkeley TCP/IP.

  I've also had someone report the message ``sendmail: accept:
  SIOCGPGRP failed errno 22'' on an SVR4 system.  This message is
  not in the sendmail source code, so I assume it is also a bug
  in the sockets emulation.  (Errno 22 is EINVAL "Invalid Argument"
  on all the systems I have available, including Solaris 2.x.)

* Sending user deletion not done properly in :include: lists.

  If you don't have the "m" (me too) option set, then a person
  sending to a list that contains themselves should not get a copy
  of the message.  However, if that list points to a :include: file
  that has one address per line, this will break, and the sender
  will always get a copy of their own message, just as though the
  "m" option were set.

  You can eliminate this by adding commas at the end of each line
  of the :include: file.

* Excessive mailing list nesting can run out of file descriptors.

  If you have a mailing list that includes lots of other mailing
  lists, each of which has a separate owner, you can run out of
  file descriptors.  Each mailing list with a separate owner uses
  one open file descriptor (prior to 8.6.6 it was three open
  file descriptors per list).  This is particularly egregious if
  you have your connection cache set to be large.

(Version 8.19, last updated 11/13/94)