COMMAND PID USER FD TYPE DEVICE SIZE/OFF
NODE NAME
perl 79714 vscan 10u IPv4 0xf80164e96b40 0 UDP
$amavis.ip:59511->$nameserver.ip:53
perl 79714 vscan 12u IPv4 0xf802071cb880 0 UDP *:47998->*:*
Very interesting. In my tests, the secon
We have encountered an issue while setting up a catchall mailbox for *@
domain.com, and I would greatly appreciate your assistance.
[...]
I think the query ends up replacing parameter %k by an address like
specific-addre...@domain.com. Since this address is not explicitly defined
in the table, no
For example, say .rar isn't supported natively. I've seen tutorials suggest
installing related packages, but don't see how Amavis picks up on that.
From the amavsid.conf, I see a "@decoders" which seems like a map
(extension, command, arguments). This appears to map to
/lib/Amavis/Unpackers.pm.
However, I don't want spam checks for the bcc-user email to be bypassed. I
also don't want the spam that would be quarantined for the regular sender
to be delivered for the bcc-user either.
In other words, I want the bcc-user to be treated exactly the same as mail
for the intended recipient. Is t
Help! I’m trying to use openDKIM with amavis-new. I have amavis up and
running with spamassassin and clamAV. That part works great, but not so
much with openDKIM.
In what way do you suspect Amavis being involved in the problem? When you use OpenDKIM, the interaction happens between Postfix
Yes, that's correct - I've always used postfix always_bcc and didn't realize
there was one for amavis as well.
If you want to keep it that way, you could declare bcc-user a spam_lover:
@spam_lovers_maps = ([
Here's a pastebin from an email similar to the above where one of the
recips is whitelisted while the other is quarantined (using report_json).
https://pastebin.com/8i6qwjvM
The paste shows
"rcpt_to": [
"bcc-u...@gambit.example.com",
"hartm...@tenney.com"
],
which is surprising, as I would exp
I was using the hash-type arrays. Would something like this work for the
hash array to represent any sender at this domain?
'.email.avi-8.com' => -100.0,
Yes, but I was under the impression that you wanted to match VERP-style
sender addresses, spec
I've just started using sender_maps instead of the welcomelist in
spamassassin directly and have a few questions. Here is my current
sender_map:
Is this `@score_sender_maps`?
'.gtenney.com' => [{ '.*@send.undone.com' => -100.0,
'.*@mg-d0.substack.
Hello amavis users,
I would like to ask those of you who have an own Amavis::Custom
implementation,
how your code interacts with Amavis internals and what your code's
assumptions are.
There is currently no convention how much or little can be changed in
Amavis::Custom context without breaking
I have in my MariaDB database used table users for other needs. Is
there any opiniion to use another table such as emails which has
structure
CREATE TABLE `emails` (
`id` int(5) NOT NULL auto_increment,
`email` varchar(255) NOT NULL,
`password` mediumtext NOT NULL,
`user_id` int(5) NO
Is this temporary fix safe to use
I have not seen feedback on Gitlab or amavis-users that it lead to problems.
until a new version of perl-DBD-MySQL is released for RHEL9 (and AlmaLinux9 and
others)?
Unless some changed their mind, there won't be a fix in perl-DBD-MySQL.
Instead you can try
Hi everybody,
I kindly ask for assistance.
Amavis doesn't get the right spam score values from policy table in
MySQL. When columns (spam_tag2_level, spam_tag3_level,
spam_kill_level) are unmodified and are NULL, the default values are
from Amavis config file. But when we modify the row, the ret
Hello amavis users,
today we have new releases in the 2.12 and 2.13 stable branches.
They address a security issue, identified as CVE-2024-28054.
See [1] or [2].
[1] https://gitlab.com/amavis/amavis/-/raw/v2.12.3/RELEASE_NOTES
[2] https://gitlab.com/amavis/amavis/-/raw/v2.13.1/RELEASE_NOTES
I'm guessing it has something to do with the interaction of
Spamassassin and Amavis? Unsure if anything is 'broken' but everything
else seems to be working as expected.
Cron test -e /usr/sbin/amavisd-new-cronjob &&
/usr/sbin/amavisd-new-cronjob sa-sync
Number found where operator expected a
when amavis rejects a message above spam kill level the reject message looks
like that:
554 5.7.0 Reject, id=27417-12 - spam (in reply to end of DATA command)
How is it possible to change the simple "spam" to something more sophisticated ;-) like
"rejected for policy reasons" or "local pol
A harder way would also be highly appreciated 🙂 I'd go some lengths for
avoiding a postgres/mariadb instance.
I suppose just whitelisting the "bayes_path" property in the user-settings
would impose a trove of new problems?
I don't know what you mean by whitelisting the property. The reason fo
Is there absolutely no way to read bayes tokens from user's home dirs using the
default berkdb files (even if it's slow)?
How would you manage file access to user's home dirs? Amavis (and
therefore Spamassassin) does not scan with root privileges.
Did anyone manage to do this?
I don't see a si
Any idea where else to dig into?
Sorry, not my area of expertise. If your build-machine builds with
`-march=x86-64` it should be fine, I guess.
Dec 26 17:03:28 batman spamd[86118]: spamd: handled cleanup of child pid [5276]
due to SIGCHLD: DIED, signal 4 (0004)
spamd has been killed with signal 4 (illegal instruction). Do you have built perl for the wrong architecture? Mail::DKIM for
example is used by both Amavis and SpamAssassin, an
amavisd-milter then does log this line:
could not read from amavisd socket /var/amavis/amavisd.sock: No error: 0
The amavisd-milter code is:
logqidmsg(mlfi, LOG_ERR, "could not read from amavisd socket %s: %s",
amavisd_socket, strerror(errno));
logqidmsg(mlfi, LOG_DEBUG, "amavisd resp
i want to bypass amavis for a specific *recipient* domain (not sender!).
I am not sure how to do that, because all i find is whitelisting a
sender.
Or should i do it better in amavisd.conf:
@bypass_spam_checks_maps = ( ['@example1.com', '@example2.com'], );
Do you want to bypass amavis comple
If there is anything hostile to mailing lists in DMARC specification,
it's this.
...
The mailing list has nothing to do with that.
Seems contradictory to me.
If you fo=1 on your domain:
You will get bombed ...
Those are the same `you`s, are they not? `you` get what `you` wished
for. If
This in my understanding generates failure reports for any forwarded
mail including any mail to lists that do not completely rewrite From:
(including this one mailing list)
- even if DKIM is preserved and valid, such mail won't generate
aligned SPF pass
unless you have better explanation of
Looking at it, fo=0 should generate dmarc report for each individual
mail forwarded, either through mailing list or via other ways.
If there is anything hostile to mailing lists in DMARC specification,
it's this.
Why would someone pick a mailing list address as their ruf?
if policy uses fo=0 then yes, both SPF and DKIM must exist, and both must pass.
if policy uses fo=1 then no, as a minimum /either/ SPF or DKIM must exist, and pass, so DMARC will work with only SPF or only
DKIM, it will also work with both, which has the advantage that only one of these must p
So Amavis is setup as an smtpd_milter as well?
No, Amavis is setup as a content_filter (content_filter =
amavis:[127.0.0.1]:10021)
You can't do that. OpenDMARC needs to see Authentication-Results for DKIM.
Also, since they allude to "some passing", I guess they did remember to set
enable_dkim_verification=1 ?
"Some passing OpenDMARC" might mean that they pass SPF-based only.
OpenDMARC is setup as a smtpd_milter in Postfix.
So Amavis is setup as an smtpd_milter as well?
Can someone maybe shed some light on why this would be happening or is there a
different way to handle DMARC?
Do you see DKIM-related Authentication-Results headers in incoming mails?
...
then I would like to quarantine this mail as spam.
...
There are two standard ways to categorize mail as spam:
- via a spam_scanner as part of @spam_scanners (usually SpamAssassin)
- via an av_scanner as part of @av_scanners and
@virus_name_to_spam_score_maps.
Other non-standard ways i
I am reading amavis documentation and would like some clarification
about the way amavis is handeling the a message score by SpamAssassin.
With a configuration like:
$sa_tag_level_deflt = 2.0; # add spam info headers if at, or above that level
$sa_tag2_level_deflt = 5.0; # add 'spam detected'
ALTER TABLE policy MODIFY spam_tag_level varchar(10) NULL;
ALTER TABLE policy MODIFY spam_tag2_level varchar(10) NULL;
ALTER TABLE policy MODIFY spam_kill_level varchar(10) NULL;
ALTER TABLE policy MODIFY spam_dsn_cutoff_level varchar(10) NULL;
ALTER TABLE policy MODIFY spam_quarantine_cutoff_leve
Could you please come up with your ideas?
Not sure, but it sounds like another case of [1]. The quickest
workaround would be to use `DBI::MariaDB` instead of `DBI::mysql` in
@lookup_sql_dsn. However it seems that perl-DBD-MariaDB is not shipped
with AlmaLinux 8. You could try [2] instead.
[
That's not how it works. I can assure you my logs have "SA dbg:” lines only for
messages which recipient address matches the @debug_recipient_maps.
It triggers a DEBUG_ONESHOT event in Amavisd-new, and when the message is fully
processed in spamassassin, Amavisd-new turns of DEBUG_ONESHOT:
you can tune Amavisd-new to activate conditional debug inside SA depending on
recipients address (that’s what I’m doing)
"$sa_debug = 1" is a global setting. Your logs should be full of "SA
dbg:" lines even for emails which do not match your @debug_recipient_maps.
It would be really nice to have a setting like sa_debug_sampling (or
sa_debug_ratio, sa_debug_sample, etc.) taking an integer value (eg. 100) that
would trigger debug logging for 1 email message every sa_debug_sampling
messages (eg. 1 message every 100 messages).
The sampling can be fuzzy, for
Fedora 38 ships Amavis 2.13.0 which switched to clamdscan in Amavis
upstream.
We have a winner... Did the package manager explain/release_note why
they made that move?
The config file change [1] comes with an explanation. The Fedora setup
does not fit the explanation completely, because it r
clamscan is slow
slow and uses much of RAM, I have disabled it by unsetting
@av_scanners_backup
in my local config file.
+1
clamdscan is the client for clamd and is fast, that would be why he
is using it.
at least in debian, amavis contacts clamd directly, not by using
clamdscan:
['Cl
Is there a fuller explanation of the virus checking options.
Amavis documentation is scattered and incomplete. There is some
information in multiple readme files and the release notes.
I want Clamsdcan to check attachments, but warn me, not either bock or
delete the attachments.
You can put
###http://www.clamav.net/ - using remote clamd scanner as a backup
['ClamAV-clamdscan', 'clamdscan',
# "--stdout --no-summary --config-file=/etc/clamd-client.conf {}",
"--stdout --no-summary --config-file=/etc/clamd.d/clamd.conf {}", <--
changed file name
[0], qr/:.*\sFOUND$/m, qr/^
Amavis still tries use nonexisting user, spamassassin seems simply
ignore bayes tests in that case.
Are you sure? Maybe the new users are just below bayes_min_ham_num and
bayes_min_spam_num.
$smtp_reason_by_ccat{&CC_SPAM} = 'id=%n - spam';
$smtp_reason_by_ccat{&CC_SPAM} = '[~[:TESTS]|SPF_FAIL|id=%n - SPF
Failure ; Please contact your administrator|spam]';
$smtp_reason_by_ccat{&CC_SPAM} = '[~[:TESTS]|MYTEST|Triggered
MYTEST-Spamassassin Rule|spam]';
I tried the following but it does
usually when amavisd-new rejects mail as pre-queue filter, the
smtp-dialog-message is like "554 5.7.0 Reject, id=13158-01 - spam"
I would like to customize the message, if the Spamain-Rule
SPF_FAIL triggers, e.g.
"554 5.7.0 Reject, id=13158-01 - SPF Failure ; Please contact your
administra
I was thinking of “fixing” this by setting always_bcc to a local alias such as
archive@localhost and then expanding this in Postfix to the two recipient
address
Sounds reasonable.
but I suspect that will automatically route the email via amavisd again.
I am sure there is a way in Postfix to
I would like to always_bcc to two different recipients, is this possible?
Surprisingly, amavis does not seem to support it.
For some reason it's trying to alter a file that doesn't exists
/var/spool/amavisd/tmp/amavis-20230110T095247-31601-RpGveeFM/email-repl.txt,
but
/var/spool/amavisd/tmp/amavis-20230110T095247-31601-RpGveeFM/email.txt
does exists
Are you sure that email-repl.txt did not exist? Amavis wants to c
Would it be possible to change URLs from 'http' to 'hXXp' for example
like most spam reporting agencies do to prevent help desk people from
clicking on malicious URLs being reported?
Out-of-the-box there is no such feature.
I added such an entry in the amavis configuration, maybe you know why
it doesn't work?
$sql_clause{'sel_wblist'} = 'SELECT "W" as wb';
Apparently the wblist code expects a user id. Try this sel_policy:
SELECT username AS id, value AS spam_kill_level FROM userpref WHERE
preference = "required_
I read it and unfortunately I still don't know how to set it up.
@lookup_sql_dsn i have set for vmail database.
Please write how to make amavis read "required_score" from database
"spamassassin" from table "userpref" and if it finds a user let it use
individual spam threshold "value".
Something
Unrelated to SQL, amavis does not care about SpamAssassin's required_score. If
you want spam-thresholds in SQL, have a look at [1].
Doesn't read required_scores value (3), only uses global value from
amavisd.conf setting: $sa_tag2_level_deflt = 1.9;
W dniu 25.11.2022 o 16:05, Da
Your user has three preferences defined, namely ok_locales, ok_languages
and required_scores.
What exactly would you like SpamAssassin to do that doesn't work?
In attachment.
W dniu 24.11.2022 o 22:04, Damian pisze:
I do not see why individual settings for users from the "userpr
I do not see why individual settings for users from the "userpref"
table are not downloaded.
What do you get for
SELECT preference, value FROM userpref WHERE username =
'mar...@wekstunters.pl' OR username = '$GLOBAL' ORDER BY username ASC
Running it on the complete email.txt (which is not how amavis invokes
SpamAssassin):
Aug 30 13:10:30.866 [73086] dbg: timing: total 7235 ms - init: 3338 (46.1%),
parse: 28 (0.4%), extract_message_metadata: 266 (3.7%), get_uri_detail_list: 56
(0.8%), tests_pri_-1000: 106 (1.5%), compile_gen:
How do I determine what exactly is wrong with the header that the
remote server is sending?
What is your log_level and do you see "check_header: " log lines?
$log_level = 1;
I don't see check_header in the logs.
What should I set it to?
$log_level = 2;
If I can't get the remote site to fix
I was wondering if anyone is seeing similar behaviour or has any
recommendations to debug this further.
Depending on your mail throughput, you could increase $log_level up to 5
and see what is being logged right before the lockup. Additionally a
mail which resides 30-40 minutes in processing
I've got a few remote senders whose messages my server is rejecting and putting this in syslog: Blocked BAD-HEADER-0
{RejectedInbound,Quarantined}
This is strange, I cannot find a code path which sets CC_BADH without setting the
minor > 0.
How do I determine what exactly is wrong with the hea
AFAIK this one should exist by default
Upstream amavis defines MYNETS in its amavisd.conf, Debian doesn't.
when submitting mail locally, I get no policy bank assigned.
I assumed that localhost is by default in @mynetworks (amavisd-new
line 931)
and the mail should get into MYNETS bank (amavisd-new line 1748)
...it obviously does not:
Dumb question: Do you even have a MYNETS policy bank?
Hi all - can I ask again if someone has any ideas on how to specify an
alternative config path for SpamAssassin files with amavis?
How about [1]?
[1] https://gitlab.com/amavis/amavis/-/blob/master/RELEASE_NOTES#L5439-5444
May 9 16:58:42 mail amavis[1542]: (01542-01) (!)IKARUS scan.server: Error reading from socket: Connection reset by peer at
/usr/sbin/amavisd-new line 8246., retrying (2)
When within a stream does that happen? Have you looked at a tcpdump?
I'm trying to find or develop a way to quarantine mail from specific
senders for specific domains. Is there a well-defined way to do this?
It appears the amavisd.conf doesn't provide the ability to specify
different policies based on domains.
Since you already use Amavis::Custom, you can amend t
Yet, I would NOT be surprised by ClamAV not detecting it, but I would
hope some third-party db (through ExtremeSHOCK) to catch it.
Does your setup permit executables?
Important question: If I provide unrar to the OS, would I need to
change anything to the amavis configuration?
Ideally, no. Look for "Found decoder for" lines on Amavis restart.
Also, would that be enough for archive scanning needs or I should try
to find 7zz too
I don't know the success rates
How about using 'unar' - a command line version of 'The Unarchiver'
(https://theunarchiver.com/) which supports a lot of formats
(including 7z, RAR, ...) and is available on several Linux distros?
Currently unsupported [1] by Amavis.
[1] https://gitlab.com/amavis/amavis/-/issues/84
You could give 7zip [1] a try. According to [2], 7zz integration does
not need code changes. You'd have to deploy 7zz manually, though, as
I don't see it in EPEL either.
RPM names are p7zip and p7zip-plugins on EPEL for Centos 8.
p7zip and p7zip-plugins do not provide 7zz, but 7z and 7za. Pe
Question 1: Is there something wrong in the configuration that does
not allow 7z scanning or probably an additional software library is
needed?
Your system lacks rar and unrar, and EPEL does not provide it any more.
Question 1A: If a decoder fails, could amavis be explicitly configured
to tr
Yet, we are recently receiving a lot of mails with virus-infected
attachments (usually in 7z files) which are scanned by amavis/clamav
but are NOT getting detected.
How do you know that they are infected? Is your setup not detecting
those viruses because neither Amavis nor ClamAV look inside
Though I can fill a bug on gitlab if you like.
Yes, please.
I would expect messages with policy POL1 to be scanned only through
SpamdClient; but what really happens is that both Spamassassin and
SpamdClient are called, and their scores added (looks like global
setting is applied).
Looks like a bug to me.
could anybody help me out to configure F-secure as virus scanner?
See [1].
[1]
https://gitlab.com/amavis/amavis/-/blob/master/conf/amavisd.conf#L554-558
The Distribution is: Centos 7.8
I don't see 2.12.2 in CentOS or EPEL. Anyway, this seems to be [1].
[1] https://gitlab.com/amavis/amavis/-/issues/85
I get the following warning message in amavis-v2.12.2. I could not find
a solution in google. What can I do to resolve the warning message.
Which distribution?
I am not sure why it is finding an "asc" file as only a PDF was attached.
The banned files mechanism checks mime parts, not only attachments.
Furthermore, even if I added ".asc" to the list above, it is still flagging the
email as having banned content.
A text-file attachment internally tran
bypass_spam_checks_maps => ['@whitelist_sender_maps'], # don't spam-check
this mail
bypass_banned_checks_maps => ['@whitelist_sender_maps'], # don't banned-check
this mail
bypass_header_checks_maps => ['@whitelist_sender_maps'], # don't header-check
this mail
Those are not per sender bu
I see no way to notify the user which attachments have been stripped.
How might this be achieved through amavis (whether through altermime
or another means)?
It might work with Amavis + Anomy::Sanitizer [1] and its "msg_file_drop"
mechanism.
[1] http://mailtools.anomy.net/sanitizer.html
https://spamassassin.apache.org/full/3.4.x/doc/Mail_SpamAssassin_Conf.html#subjprefix
Good call. But if Amavis is calling SA (which is normal) this feature
in SA can still be used can't it?
SA-in-Amavis cannot rewrite a header on its own.
apparently, our users are receiving emails containing links and /or
references to various publications, some of which are considered
'predatory'. Emails usually contain ISSN (International Standard
Serial Number) and the idea is to scan the body of an email for the
ISSN and if it matches one on
does not match $max_servers.
You mean I should change the processes number below (2) to match
amavis $max_servers configured value (currently 10)?
This would be one possible way, yes. A good number depends on cpu and
memory usage and availability. SpamAssassin can be greedy, and clamscan
(not
smtp-amavis unix - - n - 2 lmtp
maxproc does not match $max_servers.
-o smtp_data_done_timeout=1200
-o smtp_send_xforward_command=yes
These are the wrong options for an lmtp service. This is why you see:
client IP address unknown, fetched from Receive
However, it still takes time (up to 10 minutes) for final delivery,
because there is a mail processing queue in which the mail has to
enter, be examined (even minimally) and be returned. Amavis is taking
time because clamd scan is time consuming and it is processing mails
which are ahead (in qu
A suggestion was to whitelist the sender address (at the example below: no-re...@sharepointonline.com), but we fear that this
(or other similar) commonly used sender address may be deceptively used in third-party phishing/malicious mail which will then
get through unprocessed/unfiltered.
You ar
Is it possible to allow certain file types from certain addresses to
certain recipients without knowing the sending IP to add them to a
policy bank?
File types are considered in the context of banned checks, which can be
configured by various banned_* confvars. Generally some confvars are
glob
malware should be detected by clamav or other AV.
I did not understand Alex' question regarding "stop scanning" to apply
to spam-scanning only. Maybe they should specify.
Feb 24 00:00:05 [server] /usr/local/sbin/amavisd[97857]: (97857-03)
rmdir_recursively: /var/amavis/tmp/amavis-20220223T235319-97857-Mo9niBLX/parts,
excl=
# cat /var/log/amavisd.log | head -n 5
Feb 24 00:00:00 [server] newsyslog[63582]: logfile turned over
Feb 24 00:00:59 [server] /usr/local/sbi
Yeah, amavis was written at a time when 100k was a large attachment.
Have people mostly moved on from amavis? What are the other options?
I've been using it almost from the beginning (2001?), but trying to do
more with it now, and it's not holding up.
Do I understand correctly, that you want a
Feb 24 00:00:05 [server] /usr/local/sbin/amavisd[97857]: (97857-03)
rmdir_recursively: /var/amavis/tmp/amavis-20220223T235319-97857-Mo9niBLX/parts,
excl=
...
Nothing until monit(1) starts amavisd a minute later
What is (the last of) ... there?
Is there a variable to skip processing messages that
contain attachments larger than a specific size?
There isn't.
['Dr. Web Security Space ',
'drweb-ctl', 'checkmail {}',
qr/^Reject/m,,
qr/^(?:infected with:\s+(.+)/m ],
Have a look at [1].
- Both your patterns anchor at beginning of line.
- A clean mail would not match "Reject" anywhere, but you defined it for the
"clean" pattern.
[1] https://gitlab.c
I have @local_domains_maps set to all domains for which it
should process mail, but I'd like to be able to control virus/spam
scanning, blocklist/allowlist, and other features on a per-domain
basis. Is this possible?
I'm aware of $policy_bank but isn't that based on IP address? Can I
set @client_
can amavisd detect if mail was authenticated via SMTP?
Not directly.
or do I need to to explicitly set originating = 1 ?
Yes.
What's your (and other people's) advice?
You can place lines 8-13 from Savvas Karagiannidis' link in /etc/magic.
I realized that this won't help. The new definition in /etc/magic will not match (which is the purpose of the upstream change), so
file will continue and use the definition from th
What's your (and other people's) advice?
You can place lines 8-13 from Savvas Karagiannidis' link in /etc/magic.
p.path BANNED:1 us...@noa.gr: "P=p001,L=1,M=text/plain,T=exe",
matching_key="(?^:^\\.(exe|lha|cab|dll|gz|tgz|lzh)$)"
But the mail is simple text. Why is it getting dropped?
Use a $log_level of at least 4 to see the type_long that resulted in an exe
type_short, e.g.:
(3882952-01) lookup [map
how do I disable bad header checks ?
the few emails in badh inbasket all genuine, so I think I don't really
need it ?
Do you want to disable the checks or the quarantine?
I've noticed some of my emails spend up to 5 minutes in amavis checks,
dumb question, how to figure out where is the bottleneck ?
Look for TIMING log lines.
I'm running amavis 20160426 on Ubuntu 20.04 and suddenly it stopped
working. It looks like it is failing to find/read some perl modules.
Is there anybody who knows how to fix it? Thanks a lot in advanced! :)
$ ls /usr/local/lib/x86_64-linux-gnu/perl/5.30.0/
Alien auto Bit Compress
I've yet to find an example (ideally visual) of what 7bitmime and 8bitmime look
like side-by-side.
: 436f 6e74 656e 742d 5472 616e 7366 6572 Content-Transfer
0010: 2d45 6e63 6f64 696e 673a 2038 6269 740d -Encoding: 8bit.
0020: 0a0d 0ac3 a4c3 b6c3 bc0d 0a ...
we use amavisd-milter
when receiving from the world, perhaps it can cause the issue?
Ah yes, now I see it.
What does “force MTA conversion to 7-bit (e.g. before DKIM signing)” mean?
You are probably referring to
smtpd_discard_ehlo_keywords => ['8BITMIME']
If an MTA has 8BITMIME [1] mail to deliver to a next hop (e.g. an amavis instance), and this next hop does not announce being
capable of handl
noticed that Subject: is mapped to subject: (lowercase) when I define
$sa_spam_subject_tag
I am not able to reproduce this. Do you have a minimal example config?
1 - 100 of 196 matches
Mail list logo
