Package: release.debian.org
Severity: normal
Tags: buster
User: release.debian....@packages.debian.org
Usertags: pu
This is similar to #1003261 for bullseye, although there are fewer
Debian specific changes because most weren't applicable to or would have
been more invasive for buster.
I've put together my usual postfix post-release update. Because I'm
behind, it's rather larger than usual. Also, a number of packaging bugs
that apply to buster have been recently fixed in Bookworm, so the low
risk changes there have been backported too.
The diff is rather large, so I don't include it in the original bug
report to increase the chances this makes it to the mailing list.
Also, as usual, I have this update ready to upload and running in
production locally.
Here is the debian/changelog entry for this update:
postfix (3.4.23-0+deb10u1) buster; urgency=medium
[Scott Kitterman]
* Refresh patches
* Update d/p/70_postfix-check.diff to exclude makedefs.out from synlink
check. Closes: #926331
* Do not override user set default_transport in postinst. Closes: #988538
* Remove left-over ca-certificates.crt file from postfix chroot.
Closes: #991609
* Add information about keeping resolv.conf up to date in the chroot with
the resolvconf package. Closes: #964762
[Sergio Gelato]
* Correct if-up.d to not error out if postfix can't send mail yet.
Closes: #959864
[Paride Legovini]
* d/postfix.postinst: tolerate search domain with a leading dot.
Closes: #991950
[Wietse Venema]
* 3.4.15
- Bugfix (introduced: Postfix 3.0): minor memory leaks in the
Postfix TLS library, found during tests. File: tls/tls_misc.c.
- Bugfix (introduced: Postfix 3.0): 4kbyte per session memory
leak in the Postfix TLS library, found during tests. File:
tls/tls_misc.c.
- Workaround for distros that override Postfix protocol
settings in a system-wide OpenSSL configuration file, causing
interoperability problems after an OS update. File:
tls/tls_client.c, tls/tls_server.c.
* 3.4.16
- Bugfix (introduced: Postfix 3.4.15): part of a memory leak
fix was backported to the wrong place. File: tls/tls_misc.c.
- The Postfix 3.4.15 workaround did not explictly override
the system-wide OpenSSL configuration of allowed TLS protocol
versions, for sessions where the remote SMTP client sends
SNI. It's better to be safe than sorry. File: tls/tls_server.c.
* 3.4.17
- Bugfix (introduced: Postfix 3.4, already fixed in Postfix
3.6): tlsproxy(8) was using the wrong DANE macro for
connections with DANE trust anchors or with non-DANE trust
anchors (WTF: Thorsten Habich found this bug in the use
case that has nothing to do with DANE). This resulted in a
global certificate verify function pointer race, between
TLS handshakes that use TLS trust achors and handshakes
that use PKI. No memory was corrupted in the course of all
this. Viktor Dukhovni. File: tlsproxy/tlsproxy.c.
- Cleanup: the posttls-finger '-X' option reported a false
conflict with '-r'. File: posttls-finger/posttls-finger.c.
* 3.4.18
- Bugfix (introduced: Postfix 2.0): smtp_sasl_mechanism_filter
ignored table lookup errors, treating them as 'not found'.
Found during Postfix 3.6 development. File: smtp/smtp_sasl_proto.c.
- Bugfix (introduced: Postfix 2.3): when deleting a recipient
with a milter, delete the recipient from the duplicate
filter, so that the recipient can be added back. Backported
from Postfix 3.6. Files: global/been_here.[hc],
cleanup/cleanup_milter.c.
- Bugfix (introduced: before Postfix alpha): the code that
looks for Delivered-To: headers ignored headers longer than
$line_length_limit. Backported from Postfix 3.6. File:
global/delivered_hdr.c.
- Bugfix (introduced: Postfix 2.8): save a copy of the
postscreen_dnsbl_reply_map lookup result. This has no effect
when the recommended texthash: look table is used, but it
may avoid stale data with other lookup tables. File:
postscreen/postscreen_dnsbl.c.
- Bugfix (introduced: Postfix 2.2): after processing an
XCCLIENT command, the smtps service was waiting for a TLS
handshake. Found by Aki Tuomi. File: smtpd/smtpd.c.
- Bugfix (introduced: Postfix 2.3): static maps did not free
their casefolding buffer. File: util/dict_static.c.
* 3.4.19
- Feature: when a Postfix program makes a DNS query that
requests DNSSEC validation (usually for Postfix DANE support)
but the DNS response is not DNSSEC validated, Postfix will
send a DNS query configured with the "dnssec_probe" parameter
to determine if DNSSEC support is available, and logs a
warning if it is not. By default, the probe has type "ns"
and domain name ".". The probe is sent once per process
lifetime. Files: dns/dns.h, dns/dns_lookup.c, dns/dns_sec.c,
test_dns_lookup.c, global/mail_params.[hc], mantools/postlink.
- The default "smtp_tls_dane_insecure_mx_policy = dane" was
causing unnecessary dnssec_probe activity. The default is now
"dane" when smtp_tls_security_level is "dane", otherwise it is
"may". File: global/mail_params.h.
* 3.4.20
- Missing null pointer checks (introduced: Postfix 3.4) after
an internal I/O error during the smtp(8) to tlsproxy(8)
handshake. Found by Coverity, reported by Jaroslav Skarvada.
Based on fix by Viktor Dukhovni. File: tls/tls_proxy_client_scan.c.
- Null pointer bug (introduced: Postfix 3.0) and memory leak
(introduced: Postfix 3.4) after an inline: table syntax
error in main.cf or master.cf. Found by Coverity, reported
by Jaroslav Skarvada. Based on fix by Viktor Dukhovni. File:
util/dict_inline.c.
- Incomplete null pointer check (introduced: Postfix 2.10)
after truncated HaProxy version 1 handshake message. Found
by Coverity, reported by Jaroslav Skarvada. Fix by Viktor
Dukhovni. File: global/haproxy_srvr.c.
- Missing null pointer check (introduced: Postfix alpha) after
null argv[0] value. File: global/mail_task.c.
* 3.4.21
- Bugfix (introduced: Postfix 2.11): the command "postmap
lmdb:/file/name" handled duplicate keys ungracefully,
discarding entries stored up to and including the duplicate
key, and causing a double free() call with lmdb versions
0.9.17 and later. Reported by Adi Prasaja; double free()
root cause analysis by Howard Chu. File: util/slmdb.c.
- Typo (introduced: Postfix 3.4): silent_discard should be
silent-discard. File: proto/BDAT_README.html.
- Support for Postfix 3.6 compatibility_level syntax, to avoid
fatal runtime errors when rolling back from Postfix 3.6 to
an earlier supported version, or when sharing Postfix 3.6
configuration files with an earlier supported Postfix
version. File: global/mail_params.c.
* 3.4.22
- Bugfix (introduced: Postfix 3.4): the texthash: map
implementation did not support "postmap -F" behavior.
Reported by Christopher Gurnee, who also found the missing
code in the postmap source. File: util/dict_thash.c.
- Bugfix (introduced: 1999, Postfix 2.11) latent false "Result too
large" (ERANGE) errors because an strtol() call had no 'errno
= 0' statement before the call. Back-ported from Postfix 3.6.
Files: postscreen/postscreen_tests.c, util/mac_expand.c.
- Bugfix (introduced: Postfix 3.3): "null pointer read" error
in the cleanup daemon when "header_from_format = standard"
(the default as of Postfix 3.3) and email was submitted
with /usr/sbin/sendmail without From: header, and an all-space
full name was specified in 1) the password file, 2) with
"sendmail -F", or 3) with the NAME environment variable.
Found by Renaud Metrich. File: cleanup/cleanup_message.c.
(Closes: #968057)
- Bugfix (introduced: 1999): the Postfix SMTP server was
sending all session transcripts to the error_notice_recipient,
instead of sending transcripts of bounced mail to the
bounce_notice_recipient. File: smtpd/smtpd_chat.c.
- Bugfix (introduced: Postfix 2.4): false "too many reverse
jump" warnings in the showq daemon. The loop detection code
was comparing memory addresses instead of queue file names.
It now properly compares strings. Reported by Mehmet Avcioglu.
File: global/record.c.
* 3.4.23
- Bitrot: OpenSSL 3.x requires const. File: tls/tls_misc.c.
- Bugfix (bug introduced: Postfix 2.10): postconf -x produced
incorrect output, because different functions were implicitly
sharing a buffer for intermediate results. Reported
by raf, root cause analysis by Viktor Dukhovni. File:
postconf/postconf_builtin.c.
- Bugfix (problem introduced: Postfix 2.11): check_ccert_access
worked as expected, but produced a spurious warning when
Postfix was built without SASL support. Fix by Brad Barden.
File: smtpd/smtpd_check.c.
- Bugfix (introduced: Postfix 2.4): queue file corruption
after a Milter (for example, MIMEDefang) made a request to
replace the message body with a copy of that message body
plus additional text (for example, a SpamAssassin report).
The most likely impacts were a) the queue manager reporting
a fatal error resulting in email delivery delays, or b) the
queue manager reporting the corruption and moving the message
to the corrupt queue for damaged messages.
However, a determined adversary could craft an email message
that would trigger the bug, and insert a content filter
destination or a redirect email address into its queue file.
Postfix would then deliver the message headers there, in
most cases without delivering the message body. With enough
experimentation, an attacker could make Postfix deliver
both the message headers and body.
The details of a successful attack depend on the Milter
implementation, and on the Postfix and Milter configuration
details; these can be determined remotely through
experimentation. Failed experiments may be detected when
the queue manager terminates with a fatal error, or when
the queue manager moves damaged files to the "corrupt" queue
as evidence.
Technical details: when Postfix executes a "replace body"
Milter request it will reuse queue file storage that was
used by the existing email message body. If the new body
is larger, Postfix will append body content to the end of
the queue file. The corruption happened when a Milter (for
example, MIMEDefang) made a request to replace the body of
a message with a new body that contained a copy of the
original body plus some new text, and the original body
contained a line longer than $line_length_limit bytes (for
example, an image encoded in base64 without hard or soft
line breaks). In queue files, Postfix stores a long text
line as multiple records with up to $line_length_limit bytes
each. Unfortunately, Postfix's "replace body" support did
not account for the additional queue file space needed to
store the second etc. record headers. And thus, the last
record(s) of a long text line could overwrite one or more
queue file records immediately after the space that was
previously occupied by the original message body.
Problem report by Benoît Panizzon.
* Fix duplicate bounce_notice_recipient entries in postconf output.
Closes: #999694
[Andreas Hasenack]
* Fix autopkgtest so it works and uses python3
-- Scott Kitterman <sc...@kitterman.com> Fri, 07 Jan 2022 11:04:17 -0500
Scott K
