Exim 4.87 released

Richard G

Verified User
Joined
Jul 6, 2008
Messages
13,692
Location
Maastricht
This release contains the following enhancements and bugfixes:

- Internationalised mail addressing is no longer experimental
- SOCKS and PROXY-protocol proxy support are no longer experimental
- Events are no longer experimental
- REDIS support is no longer experimental

- The ACL conditions regex and mime_regex now capture substrings
- New $callout_address expansion variable
- Transports now take a "max_parallel" option
- Expansion operators ${ipv6norm:<string>} and ${ipv6denorm:<string>}
- New "-bP config" command line option
- New $dkim_key_length expansion variable
- New $prdr_requested expansion variable
- New base64d and base64 expansion items
- New main configuration option "bounce_return_linesize_limit"
- New $initial_cwd expansion variable
- The mysql support can now use msql option group names
- SOCKS proxies now support priority groups and weighted-random
selection
- REDIS lookups can now handle embedded spaces
- New event types for recipient- and host-defer
- New keep_environment and add_environment options



- Avoid trying to use OCSP-stapling on too-old GnuTLS versions
- Make logging of interfaces more consistent
- Fix problem with log files being held open
- Fix a crash post-transport on missing message-IDs
- Add Exim version string to process info
- Fix cached DNS lookups to honour TTL
- Fix transport interface settings on retries
- Ignore embedded whitespace when decoding base64
- Fix log_defer_output option of pipe transport
- Fix usage of differing hosts for different recipients of a message,
under hosts_randomize
- Fix build of Radius support
- Fix race condition in exicyclog
- Refuse unexpected arguments to exiqgrep
- Server-side advertising of TLS support now defaults to all-hosts
- Require SINGLE_DH_USE as a default, under OpenSSL
- Fix cutthrough-delivery bug on message bodies with single-dot line
- Fix spurious cutthrough delivery under the -bhc command line option
- Fix passthrough of hard-reject under cutthrough
- Fix use of DKIM private-keys having trailing '=' in the base-64
- Fix selection ordering of headers to sign for DKIM


Build-time optional feature changes:
- Extra information in DSN fail messages

The ChangeLog/NewStuff are packaged with the exim
tarball or can be reviewed online at:

http://git.exim.org/exim.git/blob/exim-4_87:/doc/doc-txt/ChangeLog
http://git.exim.org/exim.git/blob/exim-4_87:/doc/doc-txt/NewStuff
 
And I just updated via Custombuild (2.0) and having issues....

016-04-10 01:11:32 Warning: No server certificate defined; TLS connections will fail.
Suggested action: either install a certificate or change tls_advertise_hosts option
The line in exim.conf is:
tls_advertise_hosts = *

How to fix this?
 
Same here (updated with CB1.2)

Code:
2016-04-10 01:07:01 exim 4.86.2 daemon started: pid=31912, -q15m, listening for SMTP on port 25 (IPv6 and IPv4) port 587 (IPv6 and IPv4) and for SMTPS on port 465 (IPv6 and IPv4)
2016-04-10 01:10:40 Warning: No server certificate defined; TLS connections will fail.
 Suggested action: either install a certificate or change tls_advertise_hosts option
2016-04-10 01:10:40 Warning: purging the environment.
 Suggested action: use keep_environment.
2016-04-10 01:10:43 exim 4.87 daemon started: pid=5740, -q15m, listening for SMTP on port 25 (IPv6 and IPv4) port 587 (IPv6 and IPv4) and for SMTPS on port 465 (IPv6 and IPv4)


EDIT:
I think the error just relates to the 4.86.2 version of Exim, and not to v4.87.

Just got an email from my server and I think I can see in the logs that TLS is working (scroll to right to see Bold-text);

Code:
2016-04-10 02:02:54 1ap2q4-0007qC-QU => [email protected] <[email protected]> F=<[email protected]> R=lookuphost T=remote_smtp S=1713 H=mta7.am0.yahoodns.net [66.196.118.33] [b]X=TLSv1.2:ECDHE-RSA-AES128-GCM-SHA256:128[/b] CV=yes C="250 ok dirdel"
 
Last edited:
I saw that warning too after installing 4.87. But i checked everything and all the connections from the mail client to the server, and TLS connections are working fine. :confused:
 
Yep, it's indeed a strange notice. TLS connections are working fine here too. It seems we can ignore the error.
 
Back
Top