Exim has had a native SRS capability since 4.95, which would save
you a whole bunch of complexity.
On 2025/08/07 12:11 PM, Peter Thomassen via Exim-users wrote:
The added X-SRS header contains the expression
<${quote_local_part:${local_part:$address_data}}@${domain:$address_data}>
which
On 2025/08/05 4:53 PM, Peter Thomassen via Exim-users wrote:
I'd like this "mail" command to run only when the .reply.msg file exists,
The docs for exim-filters talk about all of Exim's string-expansions being
available. There's an expansion-condition "exists". Use that, in a
"${if {result-i
On 2025/08/03 4:14 PM, Andreas Metzler via Exim-users wrote:
On 2025-07-25 Moritz Orbach via Exim-users wrote:
Hi all,
I don't trust libspf2 anymore because after almost 2 years it's still
unclear to me if CVE-2023-42118 is fixed or not (e.g.
https://bugs.gentoo.org/916493#c2).
[...]
Pleas
On 2025/07/25 10:52 AM, Jeremy Harris via Exim-users wrote:
You'd need to write an alternate SPF driver module for Exim, replacing
src/miscmods/spf.c .
A bare-bones implementation is now available in HEAD. Please see the
doc-txt/experimental-spec.txt file.
I'd be interested
On 2025/07/31 10:47 AM, Jeremy Harris via Exim-users wrote:
On 2025/07/31 10:09 AM, Jeremy Harris via Exim-users wrote:
On 2025/07/29 7:38 AM, Evgeniy Berdnikov via Exim-users wrote:
Other macro definitions are displayed as expected. Similar output is
with other Exim versions. What's
On 2025/07/31 10:09 AM, Jeremy Harris via Exim-users wrote:
On 2025/07/29 7:38 AM, Evgeniy Berdnikov via Exim-users wrote:
Other macro definitions are displayed as expected. Similar output is
with other Exim versions. What's wrong?
Smells like a bug. I'll take a look.
Right
On 2025/07/29 7:38 AM, Evgeniy Berdnikov via Exim-users wrote:
Other macro definitions are displayed as expected. Similar output is
with other Exim versions. What's wrong?
Smells like a bug. I'll take a look.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## ht
On 2025/07/28 5:50 PM, Andreas Metzler via Exim-users wrote:
* changing #define PDKIM_DEFAULT_SIGN_HEADERS in
src/src/miscmods/pdkim/pdkim.h
* Updating the docs, including a proper entry for doc/doc-txt/ChangeLog
* Handling possible testsuite failures.
+ Tracking the interminable arguments o
On 2025/07/28 5:52 PM, Hardy via Exim-users wrote:
is there a way to log the smtp session of a connection? (i.e. in a way swaks
shows it from its client view.)
Exim logs stuff in normal configurations. In what way is this not what you
need?
--
Cheers,
Jeremy
--
## subscription configurati
On 2025/07/27 10:40 AM, Slavko via Exim-users wrote:
Jeremy, Heiko, please, are you willing to change this?
(copied from IRC)
it's a whole bunch of work, more than just replacing a couple of strings. I
have very little enthusiasm for it
--
Cheers,
Jeremy
--
## subscription configuration (
On 2025/07/25 1:25 PM, Niklas Schell via Exim-users wrote:
Am I missing something here? Why does exim get a result code of 255 instead of
1?
Possibly your OS does not interpret shebang (the "#! command" as the first
line of a script) within the kernel but requires a shell to do it.
The ${run.
On 2025/07/25 3:32 AM, Moritz Orbach via Exim-users wrote:
I wrote an ACL that checks SPF alignment by running spfquery.pl from
Mail::SPF. This works fine and adds the Received-SPF-header, but I don't know
how or if I can pass the SPF alignment result into Exim in order to use the
“dmarc_stat
On 2025/07/17 3:11 PM, Johnnie Adams via Exim-users wrote:
The logs don't really tell me much: A message passes through this node
several times before it dies of too many headers.
You've already learned one thing. The message was forwarded to this node
when you would rather it had not been. S
On 2025/07/15 8:37 PM, Johnnie Adams via Exim-users wrote:
To my surprise, the routing is no longer working, claiming instead
there's now a mail loop.
What on earth could cause that to happen? I'm puzzled enough that I
don't know where to begin.
Start with what it is logging.
--
Ch
On 2025/06/18 5:31 AM, Viktor Dukhovni via Exim-users wrote:
Jeremy, is there anything in Exim roughly equivalent to the Postfix
"fingerprint" security level?
With the caveat that I'm no expert on what Postfix is offering,
in Exim the server cert presented for a client connection is
available i
On 2025/06/17 9:03 PM, Martin McCormick via Exim-users wrote:
Until I can get exim4 v4.96 to be happy
with sending the outgoing messages
Assuming you mean that you're willing to accept an unverifiable
certificate, meaning that you don't really care if you're talking
to the remote you wanted to
On 2025/06/17 10:30 AM, Viktor Dukhovni via Exim-users wrote:
If this is a matter of defaults, FWIW (not much in my case as a
non-user), I recommend preferring PLAIN over LOGIN by default,
That would be up to the writer of the config - Debian in this example.
Exim is providing mechanism not pol
On 2025/06/17 4:11 AM, Viktor Dukhovni via Exim-users wrote:
I am slightly surprised Exim ended up going with LOGIN rather than the
much simpler PLAIN, perhaps this choice was made by some underlying
library, ...
The server offered both; the Exim client is configured with at least LOGIN -
but i
On 2025/06/16 11:35 PM, Martin McCormick via Exim-users wrote:
2wb5agz martin tmp $ delivering 1uRCBl-000BhD-39
R: smarthost for destination.net
T: remote_smtp_smarthost for destination.net
Transport port=25 replaced by host-specific port=587
Connecting to smtp.altice.prod.cloud.openwave.ai [66.1
On 2025/06/11 4:12 PM, White, Phil via Exim-users wrote:
What is the correct method of connecting to an sqlite3 database?
https://exim.org/exim-html-current/doc/html/spec_html/ch-file_and_database_lookups.html#SECTsqlite
"There are two ways of specifying the file. The first is by using the sql
On 2025/06/08 8:58 PM, AC via Exim-users wrote:
It can be a PTR, an A/, an MX, doesn't matter to me.
You said "rDNS lookup". That implies PTR.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exi
On 2025/06/06 10:55 AM, Viktor Dukhovni via Exim-users wrote:
250 DSN
quit
221 2.0.0 Bye
I'm not convinced that was Exim; our EHLO response always has "HELP"
as the final element. DSN, if present, would be on a continuation line.
--
Cheers,
Jeremy
--
## subscription configura
On 2025/06/06 10:25 AM, Cyborg via Exim-users wrote:
which exact arguments are passed by exim to openssl, i.e. to enforce a higher
security level or to exclude specific ciphers or exchanges.
Start with docs chapter on Main Configuration Options, section 13.
--
Cheers,
Jeremy
--
## subscrip
On 2025/05/29 5:08 PM, Rowney, Dion via Exim-users wrote:
Ps, you don't need to design, only nudge.
"Events".
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.org/
## unsubscribe (doesn't requir
On 2025/05/28 8:28 PM, Rowney, Dion via Exim-users wrote:
I am working on rebuilding a mail server that handles lists.
Not enough detail.
I moved all list data to postgres and it works great.
Management is asking if its possible to use the postgres database to store list
usage information.
On 2025/05/26 11:43 PM, Andrew C Aitchison via Exim-users wrote:
Unless you are willing and able to close the connection if/when
TLS fails, there is little benefit in disabling TLS <= 1.1
Server side:
- in MAIL ACL, deny condition = ${if !def:tls_in_cipher}.
Maybe even drop.
- in authentic
On 2025/05/27 12:01 AM, Rowney, Dion via Exim-users wrote:
headers_remove = ${if eq{${lookup pgsql \
{SELECT '1' FROM listman.lists \
WHERE \
listname = '${quote_pgsql:$local_part}' \
AND lists.options LIKE '%"reply-to-list"%' } \
}}{1} \
{Reply-to} \
On 2025/05/26 10:50 PM, Rowney, Dion via Exim-users wrote:
15:46:01.457 756913 ╰───error message: "${quote_pgsql" is not a known
operator (or a } is missing in a variable reference)
15:46:01.457 756913 ├failed to expand: ${lookup pgsql {SELECT 'Reply-to' AS
result FROM listman.lists WHERE
On 2025/05/24 6:09 PM, Slawomir Dworaczek via Exim-users wrote:
How to disable depracated protocols Tls 1 and tls 1.1 and enable only strong
protocols
a) they may already be, by your system
b) if you need to go it in your exim config, it depends on
which TLS library your exim uses. Docs ar
On 2025/05/22 10:56 AM, Frank Richter via Exim-users wrote:
What’s wrong?
Looks like a bug. Thanks for finding this.
However, do note that your add_header will be using information that
is potentially modified by the DKIM ACL. Confusion is possible.
Further, the ACL is run for each sig in th
On 2025/05/14 1:49 PM, Iman Zivver via Exim-users wrote:
As far as I can see in the route we don't have access to the original email
headers or also the whole message body to extract the headers. Is there
anyway to do so?
$message_body
--
Cheers,
Jeremy
--
## subscription configuration (requ
On 2025/05/14 9:36 AM, Cyborg via Exim-users wrote:
b) test for TLS 1.2 and 1.3
because everything else is broken & insecure.
with OPENSSL i.e. in acl_check_mail:
deny condition = ${if eq{${substr_0_6:$tls_cipher}}{TLS1.2} {0}{1}}
condition = ${if eq{${substr_0_6:$tls_cipher}}{T
On 2025/05/13 10:40 PM, Odhiambo Washington via Exim-users wrote:
So my "auth_advertise_hosts = ${if IS_MTA_PORT {}{*}}" is advertising
auth, but not checking if tls_cipher is being used, right?
Right.
My brain is a little clouded now.
What is it that I am missing towards achieving this need?
On 2025/05/13 10:19 PM, Yves Goergen via Exim-users wrote:
Not only $local_part is problematic, also $domain needs to be cleaned it seems.
Some random internet article suggested that, but I couldn't see this from the
Exim documentation.
The docs chapter on expansions, listing the $domain vari
On 2025/05/06 4:16 PM, Odhiambo Washington via Exim-users wrote:
The one for lmtp doesn't match at all while the one without (uses
dovecot-lda) matches.
Have you tried debug?
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postoriu
On 2025/05/05 9:25 PM, Yves Goergen via Exim-users wrote:
This router, however, is only used if the local_part has been successfully
looked up by a prior SQL query. So by the time we're here, I know that the
value is valid.
What I'm not sure about is whether I can use the variable or should u
On 2025/04/28 5:41 AM, Jan Kriesten via Exim-users wrote:
Anyone an idea?
Use swaks, or telnet to port 25, as a manual test for those IPs.
Try traceroute.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.li
On 2025/04/25 7:38 PM, Andy Firman via Exim-users wrote:
I love the block_hosts feature
This would be a feature of your Exim configuration file(s),
not a built-in part of Exim.
This is in my block_hosts file: 194.87.0.0/16
We don't know how your config uses this file.
Is the *194.87.207
On 2025/04/21 7:14 PM, Johnnie Adams via Exim-users wrote:
Honestly, I was flailing around, trying to mitigate the issue, but I think
that's a fool's errand. I'm abandoning trying to make this syntax work.
Understanding would be better than flailing.
The first is, if this is CVE-worthy, who t
On 2025/04/22 2:08 AM, White, Phil via Exim-users wrote:
Is this possible -- and if so, what variables should I be using?
I assume that EVENT_LOG is a macro?
The auth:fail event was only introduced in 4.97. If you are running
something earlier, it won't work.
--
Cheers,
Jeremy
--
## subsc
On 2025/04/20 3:15 PM, Sander ter Steege via Exim-users wrote:
Hi Jeremy,
On 20 Apr 2025, at 15:19, Jeremy Harris via Exim-users
wrote:
On 2025/04/20 2:06 PM, Sander ter Steege via Exim-users wrote:
This is the log I made with the debug console in Thunderbird (I replaced the
actual mail
On 2025/04/20 2:06 PM, Sander ter Steege via Exim-users wrote:
This is the log I made with the debug console in Thunderbird (I replaced the
actual mail addresses and server name):
Ah, I was assuming you were talking about actual Exim debug, as opposed to
Thunderbird.
--
Cheers,
Jeremy
--
On 2025/04/19 4:36 PM, Sander ter Steege via Exim-users wrote:
But I do not understand why the From line Thunderbird is not used, because
there is one and from the looks of it in the debugging output it is valid.
OK, we'd need to see that debug run to comment further.
--
Cheers,
Jeremy
--
On 2025/04/19 2:36 PM, Sander ter Steege via Exim-users wrote:
I do like to know the other option to solve this problem and also to know why
this works, as the documentation on the options for submission mode are rather
sparse and I could not find what this actually does except forcing exim to
On 2025/04/16 4:38 PM, Johnnie W Adams via Exim-users wrote:
10:27:42 160885 ├considering: ${lookup
ldapdn{user="CN=outer_account,OU=SVCAccounts,OU=ITS,OU=Organizations,DC=ad,DC=example,DC=com"
pass="outer_password" ldaps://
auth.example.com/OU=Service%20Accounts,DC=ad,DC=example,DC=com?dn?su
On 2025/04/16 3:13 PM, White, Phil via Exim-users wrote:
- I tried both driver = cyrus_sasl and driver=plaintext.
I am unsure of the specific differences, pro's and con's of each.
Can anyone explain?
Both drivers are capable of supporting the PLAIN mechanism. The plaintext
driver is fully b
On 2025/04/12 10:29 PM, Sander ter Steege via Exim-users wrote:
This is my first post to this list, so if I missed finding this info in a FAQ
somewhere, please forgive me.
You're in the right place to be asking.
I’m configuring a new exim server (4.98 on Almalinux 9) to replace a really old
Your internal lookup, in LDAP for "uid=foo" :-
On 2025/04/12 5:30 PM, Johnnie W Adams via Exim-users wrote:
11:25:57 134826 ├considering: ${lookup
ldapdn{user="CN=svc_domainjoin,OU=SVCAccounts,OU=ITS,OU=Organizations,DC=ad,DC=example,DC=com"
pass="password" ldaps://
auth.example.com/OU=Servic
On 2025/04/11 8:56 AM, Evgeniy Berdnikov via Exim-users wrote:
I do not know how decode this answer, and have no access to server logs,
but for me it looks like a bug in Microsoft's DC.
That seems CVE-worthy.
Meantime, the OP should write his inner lookup to never return an empty answer.
On 2025/04/10 9:31 PM, Johnnie W Adams via Exim-users wrote:
That combination should be user=foo password=foo, because that's what I'm
putting into openssl for the session.
No; your coding takes one of those foo's into ldap to get something
back, which it *then* uses in the ldapauth test.
--
C
On 2025/04/10 8:19 PM, Johnnie W Adams via Exim-users wrote:
14:16:37 121712 re-binding with user= password=foo
14:16:37 121712 Bind succeeded: ldapauth returns OK
OK, I see two possible problems here.
- The exim ldapauth condition returned true for "user= password=foo".
Is that combinati
On 2025/04/09 9:49 PM, Johnnie W Adams via Exim-users wrote:
What is strange is that I have made this work; however, I get 235
Authentication succeeded when I put in bad account information. It's almost
as though Exim is seeing the bind authentication as successful and not
caring about the
On 2025/04/10 6:29 PM, Johnnie W Adams via Exim-users wrote:
server_condition = ${if and{ \
{ !eq{}{$auth1} } \
{ ldapauth { \
user=${lookup
ldapdn{user="CN=svc_domainjoin,OU=SVCAccounts,OU=ITS,OU=Organizations,DC=ad,DC=example,DC=com"
pass="password" ldaps://
auth.example.com/
On 2025/04/10 4:43 PM, Johnnie W Adams via Exim-users wrote:
So the bind authorization is succeeding, which leaves me with the question
of why I can do a 'real' authentication with the account 'foo' and the
password 'foo':
So you ned to look at how that result is being used, with reference
to t
On 2025/04/10 3:00 PM, Johnnie W Adams via Exim-users wrote:
I don't know how much of the configuration you want to see, but here's the
lookup:
user=${lookup
ldapdn{user="CN=svc_domainjoin,OU=SVCAccounts,OU=ITS,OU=Organizations,DC=ad,DC=example,DC=com"
pass="password" ldaps://
auth.examp
On 2025/04/05 6:14 PM, Gandalf Corvotempesta via Exim-users wrote:
I have a sender that is sending bounce of emails to my debug email
with a defined subject.
As these bounce of emails most of the time lead to an auto-dos to my
destination server (both sender and recipient are mine), i need
someth
On 3/26/25 4:56 PM, David R via Exim-users wrote:
Is it tagged?
Wups, forgot the --follow-tags on the push.
Done now; thanks for the heads-up.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.or
Exim 4.98.2 is released to the public.
It addresses a use-after-free. Please read
https://exim.org/static/doc/security/CVE-2025-30232.txt
for details.
--
Cheers,
Jeremy
On behalf of the Exim Maintainers
OpenPGP_signature.asc
Description: OpenPGP digital signature
--
## subscription confi
Heads up
-
A security release of Exim is expected soon; version 4.98.2
We are planning this for 14:00 UTC tomorrow, 2025/03/26.
Distros have already been warned, and provided with the
relevant sourcecode changes,
--
Cheers,
Jeremy
On behalf of the Exim Maintainers
OpenPGP_signa
On 3/25/25 3:08 PM, Frank Förster via Exim-users wrote:
Any idea what I am doing wrong?
Assuming that the mysql_servers option value is expanded.
It is not. Look at the description in the main-config chapter
of the docs: there is no dagger-mark with the Type.
Your "-be" test is merely feeding
On 3/18/25 9:10 PM, Yves Goergen via Exim-users wrote:
I'm upgrading Exim from 4.93 to 4.97 soon and read an entry in the upgrading
documentation [1]. It says I cannot use $local_part in transports anymore and
should consider $local_part_data.
Read the docs at
https://exim.org/exim-html-4
On 3/15/25 2:13 PM, John Stevens via Exim-users wrote:
I was wondering if there is an option to exim (or a secondary program) that
would take a configuration file, resolve the conditionals and write out a new
configuration file.
exim -bP config
--
Cheers,
Jeremy
--
## subscription configu
On 3/15/25 8:06 PM, Sławomir Dworaczek via Exim-users wrote:
set acl_m_bogosity = ${run{/usr/local/bin/bogofilter -d /var/spool/bogofilter
-l -p -e -u < $message_body}}
You are assuming it is run via a shell.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## h
On 3/14/25 3:12 PM, Sander Smeenk via Exim-users wrote:
Also, when i then force a delivery on the frozen message, it does log
'signing failed: LONG_LINE' again, but delivery succeeds just fine:
Run one of these with debug enabled. This might help discover
why signing is being done.
The "bad
On 3/14/25 7:18 AM, Sławomir Dworaczek via Exim-users wrote:
bogo_router checks if the header already has bogosity, if not, the message goes
to the transport, there the message status X-Bogosity Spam Han or Unsure is
assigned
if Spam, it stops processing, if Ham or unsure, it continues processi
On 3/13/25 10:58 PM, Sławomir Dworaczek via Exim-users wrote:
hello
i have a problem with exim configuration, specifically returning from
bogo_transport to the begin routers section
theoretically command = /usr/local/exim/bin/exim -oMr bogodone -bS should
reprocess the message, but it doesn't.
On 3/10/25 4:34 PM, Patrik Peng via Exim-users wrote:
There are three different routers each having the identical "set = r_dkim_data =
..." as their last statement:
set = r_dkim_data = ${if
def:h_X-DKIM-Sign:{${readsocket{/var/run/daemon.sock}{REQUEST_DATA}{30s}{\n}{}}}{}}
Be warned: a route
On 3/10/25 3:18 PM, Patrik Peng via Exim-users wrote:
Any input would be appreciated.
No clue so far. Are there only the two mentions of r_dkim_data in
your config, or are there other uses?
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/ma
Do proper validation of that local_part.
Read the docs; there are indices.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.lists.exim.org/
## unsubscribe (doesn't require an account):
## exim-users-unsubsc
On 3/2/25 12:06 AM, MRob via Exim-users wrote:
How can domain list safe only list example.com and subdomains?
sender_domains = example.com : *.example.com
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.l
On 2/27/25 2:12 PM, satya gowtham kudupudi via Exim-users wrote:
otheru...@mydomain.com ~ $ ls -la maildirsize
-rw-rw-r-- 1 otheruser mail 0 Nov 29 11:57 maildirsize
^^
What is the significance of that prompt? Is it the current directory
of your shell? Because, if so:
On 2/25/25 8:10 PM, Johnnie W Adams via Exim-users wrote:
I've been trying to get to
it with exim -bh, but now I'm wondering: Is there any way to get past
STARTTLS? There's no -oMx option for passing a certificate, so how would I
proceed from here in the SMTP dialogue?
Give up on -bh; use a too
On 2/25/25 4:03 PM, Johnnie W Adams via Exim-users wrote:
exim -d+acl -bh 100.10.10.10.587 -omaiallo...@example.com
Try that again with -d+acl+expand .
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users.list
On 2/23/25 10:54 AM, Slavko via Exim-users wrote:
that one finds right retry
rule:
exim -brt '@|process_dmarc.py'
Retry rule: ^\N\@\|.+$\N ...
So if it works, why can't you use it?
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.or
On 18/02/2025 12:21, Sławomir Dworaczek via Exim-users wrote:
how to delivery taged message as spam to local user folder spam
eg.
if $h_X-Bogosity: MATCH "Spam" ...
Have the transport that does local delivery do that test, in choosing
where to place each message. Details depend on what it curr
On 13/02/2025 01:49, Ian Z via Exim-users wrote:
exim -be '${eval10:${extract{1}{.}{11.22.33.44}}<<24 +
${extract{2}{.}{11.22.33.44}}<<16}'
# exim -be '${eval:${extract{1}{.}{11.22.33.44}}<<24 +
${extract{2}{.}{11.22.33.44}}<<16}'
-4611686018427387904
# exim -be '${eval:(${extract{1}{.}{11.22
On 09/02/2025 22:29, Daniel Botting via Exim-users wrote:
My understanding of tainting and experience so far is that it will stop your
configuration from working until you resolve the tainted data?
No.
Some data is tainted. Basically, if it came from the wire.
This is normal.
What is not per
On 07/02/2025 11:26, Olaf Hopp (SCC) via Exim-users wrote:
Is there a way to "warm up" the IP-reputation of smarthost-B by starting
with e.g. a 90:10 distribution and increase that over the time ?
The "hosts" option argument is expanded. Write a string-expansion
that returns a single value cho
On 05/02/2025 10:34, Andrew Bernard via Exim-users wrote:
Can anybody explain to me what tainting is? I find no reference to it in the
Hazel book on Exim. Consequently I have no idea how to debug issues with the
pipe driver.
Exim has rich facilities for interpreting strings.
These are known a
On 03/02/2025 07:48, Timo Sirainen via Exim-users wrote:
It attempted to preserve backwards compatibility by checking client-provided
VERSION first before sending data that the client wouldn't handle correctly.
Is there documentation available which specifies, for both new and older
versions
On 31/01/2025 17:08, Matteo Bonora via Exim-users wrote:
This transport is really very simple and the bonding is done by the "interface =
$interface_address" line. It works perfectly in every case but the one I described.
Commit 237b2df13410 looks relevant.
--
Cheers,
Jeremy
--
## subscript
On 31/01/2025 14:38, Jeremy Harris via Exim-users wrote:
On 31/01/2025 11:35, Andrew C Aitchison via Exim-users wrote:
Is anyone using Exim with OpenSSL v1.1.1 or earlier ?
There's a buildfarm system running 1.0.2g (Solaris 10)
and another running 1.0.2t (Solaris 11).
Any informant
On 31/01/2025 11:35, Andrew C Aitchison via Exim-users wrote:
Is anyone using Exim with OpenSSL v1.1.1 or earlier ?
There's a buildfarm system running 1.0.2g (Solaris 10)
and another running 1.0.2t (Solaris 11).
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## htt
On 31/01/2025 10:08, Matteo Bonora via Exim-users wrote:
Splitting the messages in different queues is interesting though. Could you
point me to some documentation on how to implement that?
https://exim.org/exim-html-4.93/doc/html/spec_html/index.html
Concept index. Named queues, and the "qu
On 31/01/2025 09:30, Matteo Bonora via Exim-users wrote:
I'm open to alternatives ;)
Well, you've not said what version Exim you are running so I
can only guess it's something modern. You could split out
to separate named-queues during reception, and have separate
queue-runners. Or you could
On 30/01/2025 18:34, Matteo Bonora via Exim-users wrote:
Is there a way to disable or control this behavior?
Try setting "connection_max_messages = 1" on the transport.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/list
On 30/01/2025 17:16, Patrik Peng via Exim-users wrote:
if there could be some issue in exims libspf2 integration or handling of the
$spf_header_comment variable.
The string for that variable is provided by the library call
SPF_response_get_header_comment().
--
Cheers,
Jeremy
--
## subscript
On 30/01/2025 15:48, Patrik Peng via Exim-users wrote:
error in processing during
The Exim source doesn't seem to include those words, so I assume they are
coming from the library (libspf2). You could possibly fake up a suitable
testcase, knowing the domain involved (which I assume you deliber
On 23/01/2025 17:35, Brian Spraker via Exim-users wrote:
I have whitelisted IP addresses before.. and also whitelisted email addresses.
However, is there a way of whitelisting an email address only if it comes from
a specific IP address?
Yes, it's logically an AND of two conditions.
Syntax wil
On 22/01/2025 12:35, Slavko via Exim-users wrote:
IMO, it can work, but will be less than optimal
Yeah, abusing the retry mech for this (I assume the
delay is to permit manual inspection of outbound messages)
is not wonderful.
I'd be tempted to divert the message to an alternate named-queue.
On 19/01/2025 21:51, Randy Bush via Exim-users wrote:
this suddenly appeared on multiple system.
Could it be that all these systems had updates applied at about the
time of it starting? If so, that's a useful commonality to start
an investigation with.
--
Cheers,
Jeremy
--
## subscription
On 16/01/2025 15:59, Randy Bush via Exim-users wrote:
what happened? the symptom was
2025-01-13 06:25:27 1tXDt8-00Aigw-35 1tXDt8-00Aigw-35 no recipients found in
headers
Exim is being fed a message useing a command-line, with a "-t" option. This
means
that recipients are to be extracted fr
On 10/01/2025 05:31, Andrew C Aitchison via Exim-users wrote:
In that case, can and should we make exim-as-client report
why it rejected the server's certificate in the main log ?
The operational problem would then become the amount of cruft
in the log, were every possible item of information r
On 09/01/2025 09:49, Gandalf Corvotempesta via Exim-users wrote:
Il giorno gio 9 gen 2025 alle ore 10:40 Jeremy Harris via Exim-users
ha scritto:
We don't know that the client end actually saw that, because you've
not shown us the debug from the client.
That's what i'
On 09/01/2025 09:01, Gandalf Corvotempesta via Exim-users wrote:
The log above should show that auth is advertised from the smarthost,
so the client should authetnicate
Actually, we only know that the smarthost transmitted an advertisement
of AUTH (and, insecure methods in cleartext; a poor
On 08/01/2025 12:07, Gandalf Corvotempesta via Exim-users wrote:
This is the smarthost
LOG: MAIN
TLS error on connection from 4.3.2.1.bc.googleusercontent.com
[1.2.3.4] (recv): A TLS fatal alert has been received: Certificate is
bad
The smarthost received an alert fron the client, saying
On 08/01/2025 11:30, Gandalf Corvotempesta via Exim-users wrote:
Any idea how to enable a full-debug only for requests coming from a
particular ip ?
There is an ACL modifier for enabling debug. Combine that with an
ACL condition selecting the source IP, in an ACL verb called from
the acl_smtp_c
On 07/01/2025 11:09, Gandalf Corvotempesta via Exim-users wrote:
As client, the smarthost server deny the email due to "A TLS fatal
alert has been received: Certificate is bad"
That's not quite clear. Is your exim saying that it received the
fatal-alert, or is your smarthost saying that?
We n
On 05/01/2025 21:27, Sebastian Nielsen via Exim-users wrote:
What I have understand, header_syntax just validates the syntax of all headers
are valid MIME, without actually checking if each header follows
header-specific rules.
https://exim.org/exim-html-current/doc/html/spec_html/ch-access_c
On 02/01/2025 20:26, Alain D D Williams via Exim-users wrote:
How do I get this to work ?
Have a look at section 4 of the docs chapter on the pipe transport.
--
Cheers,
Jeremy
--
## subscription configuration (requires account):
## https://lists.exim.org/mailman3/postorius/lists/exim-users
1 - 100 of 490 matches
Mail list logo
