Re: Daemon restarts randomly

On Mon, 2023-04-03 at 18:18 +0930, Stephen Davies wrote: > > Apr 2 03:09:57 mustang amavis[1189]: sd_notify (no socket): > STOPPING=1\nSTATUS= > Server rundown, notifying child processes. When my amavisd restarts at 3am, it's because the sa-update cron job just updated spamassassin's rules.

Re: clamd socket timeout and failover

On Thu, 2021-08-19 at 12:09 +0200, Matus UHLAR - fantomas wrote: > Hello, > > it looks that when amavisd receives timeout from clamd, it falls to calling > clamscan. > > Can I tune this timeout and optionally disable fallback? > > when our system is overloaded, calling clamscan makes it even wor

Re: clamav (under amavis) not filtering out viruses!

On 2020-10-15 08:11, Matus UHLAR - fantomas wrote: > > on debian, you need to add user clamav in amavis group and restart clamd, so > clamd can > scan files unpacked by amavis. > I suggest using the "clamdscan --fdpass" entry instead of mucking with the system users/groups. This is now the upst

Re: clamav (under amavis) not filtering out viruses!

On 2020-10-15 08:48, Dino Edwards wrote: > Hi Niko, > > Try this: > > https://github.com/extremeshok/clamav-unofficial-sigs > > This has worked wonderfully for us. > That script comes packed with silly root exploits. You'll be safer with fangfrisch: https://github.com/rseichter/fangfrisch

Re: Permission Denied Issue in amavisd service

On 2020-08-26 09:19, Rob Sterenborg (Lists) wrote: > On 26-08-2020 04:27, Aung Ye Kyaw wrote: >> *Dear Sirs,* >> >> I would like to ask one issue that I added a new dkim in amavisd.conf >> file and then restarted amavisd service but service failed so I changed >> back to original amavisd.conf fil

Re: Do not reject email for clamav Heuristics.OLE2.ContainsMacros event

On 2/19/20 9:05 AM, Patrick Proniewski wrote: > > I've tried the solution exposed in > > but I'm not using any policy_bank on my MX servers and I failed to adapt the > config snippet to make it work. > You don't need

Re: Extra support for some archive types

On 8/9/19 4:39 PM, korsar...@gmail.com wrote: > 1) ACE has no support at all, need to add it > 2) gzip - gzip -d command changes original file names of extracted files, > so amavis can't filter them. Lie you have "script.js" in "archive.gz" and > after extraction you get "archive" file, not "sc

Re: whitelist

On 7/15/19 8:09 AM, Dino Edwards wrote: >>P.S. >>Any pre-queue process will introduce a noticable delay. This is imposed > by the scan process itself. It is the same delay you have in post-queue > – its just that now you get to "see" in SMTP sessions. Clients won't > bother. The typical client time

Re: logwatch reporting Unmatched Entries

On 2/12/19 1:58 PM, Robert Moskowitz wrote: > This is a new, test, build on CentOS7, amavis-new 2.11.1 > > every day I see the following in the logwatch report > > ... > > Are any of these something I should/could something about? > Logwatch is using the "logreporters" project, http://logrep

Re: Update to SA cf requires amavis restart?

On 2/12/19 12:30 PM, Curtis Maurand wrote: > Sorry, > > Sent this directly to Dominic.  For the benefit of the list: > > sudo /usr/sbin/amavisd-new reload > The systemd service should be able to do this for you, e.g. ExecReload=/usr/sbin/amavisd -c /etc/amavisd.conf reload if the "forking"

Feature request: log read failures on configuration files

This comes up every once in a while, most recently in https://bugs.gentoo.org/646330 When amavisd-new reloads after dropping privileges, it reads its configuration files as the unprivileged "amavis" (or whatever you've set it to) user. Since that differs from the first time amavisd-new is start

Re: amavis-mc creates its PID file after dropping privileges

On 09/14/2017 10:52 AM, Michael Orlitzky wrote: > I noticed that the amavis-mc daemon creates its PID file after dropping > privileges: > > if (defined $daemon_user) { > drop_priv($daemon_user,$daemon_group); > } > > if (defined $pid_file && $pid_fil

amavis-mc creates its PID file after dropping privileges

I noticed that the amavis-mc daemon creates its PID file after dropping privileges: if (defined $daemon_user) { drop_priv($daemon_user,$daemon_group); } if (defined $pid_file && $pid_file ne '') { my $pid_file_fh = IO::File->new; $pid_file_fh->open($pid_file, O_CREAT|O_WRONLY, 0

Re: Stopping backscatter with D_REJECT

On 04/17/2017 09:00 AM, Todd D. Taft wrote: I was hoping to avoid dropping messages on the floor for well-behaved servers. While it's rare, I have found the occasional ham that gets marked as spam. With D_REJECT, at least the sender gets a notification that the message didn't go through. You

Banned files in a policy bank

I've got two MXes now, with the second using amavis on localhost:10026. I've set up a policy bank: $policy_bank{'mx1.example.com'} = { banned_filename_maps => ['mx1.example.com'], }; $interface_policy{'10026'} = 'mx1.example.com'; And the associated banned rules: %banned_rules = (

Per-domain banned extensions with pre-queue filter

We have a pre-queue filter set up with a pretty boring config. In postfix, smtpd_proxy_filter = localhost:10024 Then in amavis, we do whatever spam and virus scanning we need to do. If the message is spam/virus, we reject it at SMTP time rather than bouncing it. Anything else we pass. We don't

Re: 2.9.1 reload crashes

On 07/04/2014 12:03 PM, Nick Edwards wrote: > Today, upgraded 2.9.0 to 2.9.1, and found that when sa-update script > runs, which at the end of all of that, calls amavisd reload > > amavisd exits! > only log message is: > > (!)Net::Server: 2014/07/01-00:12:31 Re-exec server during HUP > > this ha

Re: local_domains_maps

On 01/30/2014 08:47 PM, Nick Edwards via amavis-users wrote: > Hi again > > Is there a way to have the spam check header added to every message, > regardless of @local_domains_maps setting, we host thousands of > domains, and it is not feasible to add/del domains from this setting > all the time.

Re: amavisd stops after reload

On 06/11/2013 05:19 PM, Martin Fuxa wrote: > after upgrade from 2.6.6 to 2.8.0 stops working > amavisd reload > > same with 2.8.1-rc1 > > On terminal prints > Signalling a SIGHUP to a running daemon [4756] > > in log > Jun 11 23:12:53 anta3 amavis[4756]: (!)Net::Server: 2013/06/11-23:12:53 > HU

Re: amavisd reload fails

On 03/21/2013 07:18 PM, psychobyte wrote: > Hi, > > I'm using amavisd-new-2.7.0. When ever i try "amavisd reload" the > amavisd dies and never comes back up. The logs don't seem to reveal > much detail that's useful to me. I have made sure that all config files > are readable by the amavis gro

Re: why is Mail::DKIM not optional any more in amavisd ?

On 08/01/12 10:14, Michael Orlitzky wrote: > > This is the default Gentoo conf: > > $ sudo grep dkim /etc/amavisd.conf > $enable_dkim_verification = 0; # enable DKIM signatures verification > $enable_dkim_signing = 0;# load DKIM signing code, keys defined > by d

Re: why is Mail::DKIM not optional any more in amavisd ?

On 08/01/12 08:24, Mark Martinec wrote: >> why is Mail::DKIM not optional any more in amavisd ? >> https://bugs.gentoo.org/show_bug.cgi?id=429106 >> in gentoo i have not tryed 2.8.x yet > > It is still optional, amavisd 2.7.* or 2.8.0 will > start normally without Mail::DKIM, unless you require

Silent crash on reload with insufficient permissions

(This is really a feature request, not a bug report.) I messed up our permissions last night, and prevented amavis from reading /etc/postfix/maps/rfc_addresses. In the amavis config, I have, @virus_lovers_maps = ( read_hash("/etc/postfix/maps/rfc_addresses") ); @spam_lovers_maps = ( read_has

Re: Adding support for mdb as alternative

On 10/26/11 00:41, Quanah Gibson-Mount wrote: > The OpenLDAP project has for many years used BerkeleyDB as its high > performance back end database. However, for a number of reasons, it has > been found to be less than optimal. In response, the OpenLDAP primary > developer wrote new database softw

Re: pilot error? or idiots at microsoft?

On 08/12/2011 07:53 PM, Jo Rhett wrote: > On Aug 12, 2011, at 7:49 AM, Mark Martinec wrote: >>> in fact, any connection to amavis from 169* would be strange... >>> unless your laptop also did not get a good ip and pulled a 169* >>> address. >> >> Yes. It would be unusual, although not impossible.

Re: pilot error? or idiots at microsoft?

On 08/10/11 10:48, Michael Scheidell wrote: > On 8/10/11 10:33 AM, Michael Orlitzky wrote: >> On 08/10/11 10:26, Michael Scheidell wrote: >>> so, what brain decided it would be ok to use 169.* addresses for their >>> internal ip's? >>> >>> was it

Re: pilot error? or idiots at microsoft?

On 08/10/11 10:26, Michael Scheidell wrote: > so, what brain decided it would be ok to use 169.* addresses for their > internal ip's? > > was it microsoft? (var says that ms uses these for their internal > clustering ip's for clustered exchange servers http://en.wikipedia.org/wiki/Link-local_addr

Fwd: mx1 amavis-logwatch

Probably a stupid question, but is there any way I can avoid running the backup scanner if the primary scanner timed out / got killed due to $child_timeout? I feel like clamscan is guaranteed to time out if the clamd scan did, but I would still like it to kick in if I somehow break clamd (happened

Re: Reload not working after installing 2.7.0

On 07/08/11 10:57, Bob Mortimer wrote: > > I'd changed the permissions of amavisd.conf to 640 as detailed in the > INSTALL document. Changing it back to 644 sorted it out and all is > working fine now. You should probably 'chgrp amavis' it, instead, assuming that amavisd is running as group amavi

Re: Perl warnings after 2.7.0 upgrade

On 07/06/11 10:12, Mark Martinec wrote: > Michael, > >> Probably harmless, but I racked up about 450 of these overnight after >> the upgrade to 2.7.0: >> >>Jul 4 09:20:26 mx1 amavis[26145]: (26145-08) _WARN: Argument "x" >>isn't numeric in sprintf at /usr/sbin/amavisd line 13126, >>li

Re: Perl warnings after 2.7.0 upgrade

On 07/04/2011 05:08 PM, Michael Scheidell wrote: > On 7/4/11 9:30 AM, Michael Orlitzky wrote: >> Probably harmless, but I racked up about 450 of these overnight after >> the upgrade to 2.7.0: >> >> Jul 4 09:20:26 mx1 amavis[26145]: (26145-08) _WARN: Argument "x&q

Perl warnings after 2.7.0 upgrade

Probably harmless, but I racked up about 450 of these overnight after the upgrade to 2.7.0: Jul 4 09:20:26 mx1 amavis[26145]: (26145-08) _WARN: Argument "x" isn't numeric in sprintf at /usr/sbin/amavisd line 13126, line 390.

Reasonable $child_timeout values (pre-queue)

(I just got done switching from post-queue to a pre-queue setup.) Is ~45 seconds still the accepted "reasonable" value for $child_timeout in a pre-queue setup? And, how does its value interact with Postfix's smtp_data_done_timeout? Mark once posted that $child_timeout should be less than $smtp_da