default before adding the WITH_MUTT_IMAP and WITH_MUTT_POP knobs).
- fix pkg-plist if not build with WITH_MUTT_IMAP or WITH_MUTT_POP.
- rename internal variable to less confusing name
Submitted by: Udo Schweigert <ust@cert.siemens.de>
Sendmail, Inc., and the Sendmail Consortium announce the availability
of sendmail 8.11.6 and 8.12.0.Beta19.
These new versions fix a security problem reported by SecurityFocus
regarding command line processing. This vulnerability is present in
sendmail open source versions between 8.10.0 and 8.11.5 as well as all
8.12.0.Beta versions. Therefore, sendmail 8.12.0.Beta users should upgrade
to 8.12.0.Beta19.
The problem was not present in 8.10 or earlier versions. However, as
always, we recommend using the latest version. Note that this problem is
not remotely exploitable.
- Fixes bug where setting local part to /dev/null without setting
file_transport caused Exim to crash.
- Other fixes do not apply to FreeBSD.
* Now that RELENG_4 has bzip2 in the base system, use the bz2
distfiles.
o Remove redundant BUILD_DEPENDS (only need to depend on p5-Net once).
o Add USE_PERL5.
o Sort pkg-plist.
o Don't fail on deinstall if other Mail:: modules are installed.
PR: ports/29663
Submitted by: Pete Fritchman <petef@databits.net>
* August 6 2001 -- vbsfilter-1.7
* Various fixes for the case where headers and/or body are empty.
* Patch sent in by Sergiy Zhuk <serge@yahoo-inc.com>
*
* Update "dangerous extension" list.
* Noted by David F. Russell <David.F.Russell@ncmail.net>
*
* July 25 2001 -- vbsfilter-1.6
* Also recognise 'name=blah' as well as the more usual 'name="blah"'.
* Patch sent in by Andrea Adams <andrea@vividimage.com>
*
* July 9 2001 -- vbsfilter-1.5
* Add support to reject arbitrary extensions.
*
* March 31 2001 -- vbsfilter-1.4
* Update for sendmail-8.12
*
o Update port to version 1.11
o Support NOPORTDOCS, deploy DOCSDIR
o Update DESCR, add WWW tag
PR: 29368
Submitted by: Pete Fritchman <petef@databits.net>
been broken by the incompatibility between db3.2 and db3.3.
I am unconditionally bumping the PORTREVISION's to eliminate possible
troubles that should not happen before the coming release.
Sorry for inconvenience. I'll try to check compatibility hard before
committing next time.
Fix a possible race condition when sending a HUP signal to restart
the daemon. This could terminate the current process without
starting a new daemon. Problem reported by Wolfgang Breyha
of SE Netway Communications.
Only apply MaxHeadersLength when receiving a message via SMTP or
the command line. Problem noted by Andrey J. Melnikoff.
When finding the system's local hostname on an IPv6-enabled system
which doesn't have any IPv6 interface addresses, fall back
to looking up only IPv4 addresses. Problem noted by Tim
Bosserman of EarthLink.
When commands were being rejected due to check_relay or TCP
Wrappers, the ETRN command was not giving a response.
Incoming IPv4 connections on a Family=inet6 daemon (using
IPv4-mapped addresses) were incorrectly labeled as "may be
forged". Problem noted by Per Steinar Iversen of Oslo
University College.
Shutdown address test mode cleanly on SIGTERM. Problem noted by
Greg King of the OAO Corporation.
Restore the original real uid (changed in main() to prevent
out of band signals) before invoking a delivery agent.
Some delivery agents use this for the "From " envelope
"header". Problem noted by Leslie Carroll of the
University at Albany.
Mark closed file descriptors properly to avoid reuse. Problem
noted by Jeff Bronson of J.D. Bronson, Inc.
Setting Timeout options on the command line will also override
their sub-suboptions in the .cf file, e.g., -O
Timeout.queuereturn=2d will set all queuereturn timeouts
to 2 days. Problem noted by Roger B.A. Klorese.
CONFIG: Fix parsing for IPv6 domain literals in addresses
(user@[IPv6:address]). Problem noted by Liyuan Zhou.
