Re: re-route mails on demand during block of ip address

On Tue, Mar 10, 2020 at 05:18:01PM -0500, Noel Jones wrote: > On 3/10/2020 4:55 PM, Stefan Bauer wrote: > > > > i know this is quite old, but > > smtp_fallback_relay should only get triggered on > > undeliverable-events and not when remote replies with a 4xx or 5xx - > > right? Not quite.

Re: Mail loops back to itself ONLY when using `mail` command on server

On Wed, Mar 11, 2020 at 09:50:42AM +1000, Jay wrote: > But, it seems like the true problem is finding out why the `mail` > command isn’t putting messages into > /Library/Server/Mail/Data/spool/maildrop instead of > /var/spool/postfix/maildrop. The mail(1) command undoubtedly invokes sendmail(1)

Re: Mail loops back to itself ONLY when using `mail` command on server

> On 11 Mar 2020, at 3:33 am, Noel Jones wrote: > > On 3/10/2020 12:03 AM, Jay wrote: >>> On 10 Mar 2020, at 2:02 pm, Noel Jones wrote: >>> >>> On 3/9/2020 9:09 PM, Jay wrote: Hi folks, I’m having a strange situation on one of my macOS X Servers (10.13.6) running postfix

Re: re-route mails on demand during block of ip address

On 3/10/2020 4:55 PM, Stefan Bauer wrote: Hi Noel, i know this is quite old, but smtp_fallback_relay should only get triggered on undeliverable-events and not when remote replies with a 4xx or 5xx - right? That's what the docs say, which implies my off-the-cuff crappy workaround is

Re: Client cert based relaying setup not working

On Tue, Mar 10, 2020 at 03:33:44PM +0100, Adam Cecile wrote: > submission inet  n   -   y   -   -   smtpd >     -o syslog_name=postfix/submission >     -o smtpd_tls_security_level=encrypt >     -o smtpd_sasl_auth_enable=yes > -o smtpd_tls_fingerprint_digest=sha1 >    

Re: t/s outbound 99% timeout on TBird?

On Tue, Mar 10, 2020 at 10:53:02PM +1100, li...@sbt.net.au wrote: > Mar 10 00:10:50 geko postfix/cleanup[9389]: C099F42B0143: > message-id=<5262b5f4-18d5-b7fb-b09a-be37f1d61b9e@a_tld.com> > Mar 10 00:12:38 geko opendkim[1322]: C099F42B0143: DKIM-Signature field > added (s=default, d=a_tld.com) >

Re: Warning about non-existent MX for destination domain

On Tue, Mar 10, 2020 at 12:18:28PM -0700, Fred Morris wrote: > NOERROR doesn't necessarily mean that there was an answer. The number of > answers to the question asked can be 0, even if the FQDN (fully qualified > domain name) exists, if some other type of information is associated with > it.

Re: postfix and systemctl

On 11/03/20 1:10 am, Emanuel Postfix wrote: In the file /etc/yum.repos.d/gf.repo add this config: Get rid of that, the gf-release package contains that file. from console I execute the following: yum --nogpg install

Re: Warning about non-existent MX for destination domain

Let's try to clear up DNS even though that's not what the original querant is asking. On Tue, 10 Mar 2020, Wietse Venema wrote: Catalin Bucur: On 10/03/2020 17:16, Wietse Venema wrote: This is a problem with your DNS resolver WHICH IS NOT PART OF POSTFIX. You need to use a better DNS

Re: Warning about non-existent MX for destination domain

On 10/03/2020 19:49, Dominic Raferd wrote: My responses seem like OP's: # host -t mx imake.ro 8.8.8.8 Using domain server: Name: 8.8.8.8 Address: 8.8.8.8#53 Aliases: Host imake.ro not found: 2(SERVFAIL) # host -t mx

Re: Warning about non-existent MX for destination domain

On Tue, 10 Mar 2020 at 17:37, Jaroslaw Rafa wrote: > Dnia 10.03.2020 o godz. 19:33:14 Catalin Bucur pisze: > > Google dns says: > > # host -t mx imake.ro 8.8.8.8 > > Using domain server: > > Name: 8.8.8.8 > > Address: 8.8.8.8#53 > > Aliases: > > Host imake.ro not found: 2(SERVFAIL) > >

Re: Warning about non-existent MX for destination domain

> Gerald and Bill, you have both of you right, delay_warning_time it's the > parameter I was looking for. > Anyway, after setting up delay_warning_time=5m I noticed some warnings in > logs: > > Mar 10 17:52:02 mail postfix/bounce[7326]: warning: [built-in]: zero result > in delay template

Re: Warning about non-existent MX for destination domain

On 10/03/2020 19:33, Jaroslaw Rafa wrote: If you get reply 2(SERVFAIL) it means DNS server for the domain failed and did not return any usable reply, so we actually don't know if the domain exists or not. In that case Postfix SHOULD keep trying to deliver mail (maybe the DNS server for the

Re: Warning about non-existent MX for destination domain

Dnia 10.03.2020 o godz. 19:33:14 Catalin Bucur pisze: > Google dns says: > # host -t mx imake.ro 8.8.8.8 > Using domain server: > Name: 8.8.8.8 > Address: 8.8.8.8#53 > Aliases: > Host imake.ro not found: 2(SERVFAIL) Something's wrong with resolver library on your machine. Mine says: raj@rafa:~$

Re: Warning about non-existent MX for destination domain

Dnia 10.03.2020 o godz. 19:11:33 Catalin Bucur pisze: > You mean that I should modify standard configuration for my dns > server (resolver) to force reply that MX record does not exist, > instead of "Host doman.com not found: 2(SERVFAIL)" ? If you get reply 2(SERVFAIL) it means DNS server for the

Re: Mail loops back to itself ONLY when using `mail` command on server

On 3/10/2020 12:03 AM, Jay wrote: On 10 Mar 2020, at 2:02 pm, Noel Jones wrote: On 3/9/2020 9:09 PM, Jay wrote: Hi folks, I’m having a strange situation on one of my macOS X Servers (10.13.6) running postfix 3.1.1. The server is configured to only use virtual_domains and virtual. When an

Re: Warning about non-existent MX for destination domain

On 10/03/2020 19:26, Wietse Venema wrote: SERVFAIL (for a domain without MX record) means that something is broken. If the fault is not with your local resolver, then you need make sure that you point your resolver at a proper implementation. I have seen nonsense results from cheap routers that

Re: Warning about non-existent MX for destination domain

SERVFAIL (for a domain without MX record) means that something is broken. If the fault is not with your local resolver, then you need make sure that you point your resolver at a proper implementation. I have seen nonsense results from cheap routers that expect queries from web clients that never

Re: Warning about non-existent MX for destination domain

On 10/03/2020 17:55, Wietse Venema wrote: THAT is a DNS resolver issue. If a domain has no MX record the resolver should say sa IMMEDIATELY, instead of forever returning NO RESPONSE. Again, with proper resolver will IMMEDIATELY reply that the MX record does not exist, allowing Postfix to

Re: Warning about non-existent MX for destination domain

On 10/03/2020 17:57, Gerald Galster wrote: In case there is no MX record the mail is delivered to the domain's ip (a/ record). It's up to the domain's administrator to set a NULL MX to prevent that, but it's quite new: https://tools.ietf.org/html/rfc7505 What postfix can do:

Re: Warning about non-existent MX for destination domain

> On 10/03/2020 17:16, Wietse Venema wrote: >> This is a problem with your DNS resolver WHICH IS NOT PART OF >> POSTFIX. You need to use a better DNS resolver. >> >> With a properly functioning resolver: >> >> $ host -t mx www.postfix.org >> www.postfix.org

Re: Warning about non-existent MX for destination domain

Catalin Bucur: > On 10/03/2020 17:16, Wietse Venema wrote: > > This is a problem with your DNS resolver WHICH IS NOT PART OF > > POSTFIX. You need to use a better DNS resolver. > > > > With a properly functioning resolver: > > > > $ host -t mxwww.postfix.org > > www.postfix.org has no MX record >

Re: Warning about non-existent MX for destination domain

On 10/03/2020 17:16, Wietse Venema wrote: This is a problem with your DNS resolver WHICH IS NOT PART OF POSTFIX. You need to use a better DNS resolver. With a properly functioning resolver: $ host -t mxwww.postfix.org www.postfix.org has no MX record Wietse You didn't understood my

Re: Warning about non-existent MX for destination domain

On 10 Mar 2020, at 10:26, Catalin Bucur wrote: Hello, I have a question about how Postfix generate warning messages about non-existent domains, non-existent MX domain names or for timeout connecting to a destination mail server. In fact, my real problem is that postfix does not warn me

Re: Warning about non-existent MX for destination domain

Catalin Bucur: > For an existing domain but without MX, I receive the error /"Host or > domain name not found. Name service error for name=onedomain.com > type=MX: Host not found, try again)"/, but AFTER > /maximal_queue_lifetime/, which is very annoying. In this case the > sender finds out

Re: postfix3 with opendkim

SysAdmin EM: > grep milter /etc/postfix/master.cf > # -o milter_macro_daemon_name=ORIGINATING > # -o milter_macro_daemon_name=ORIGINATING > > I can't find the problem flaw To find out what Postfix really uses: $ postconf -n | grep milter $ postconf -P | grep milter Postfix does not use

Client cert based relaying setup not working

Hello, Here is my submission definition on *server* master.cf: submission inet  n   -   y   -   -   smtpd     -o syslog_name=postfix/submission     -o smtpd_tls_security_level=encrypt     -o smtpd_sasl_auth_enable=yes #    -o smtpd_tls_fingerprint_digest=sha1 #    -o

Warning about non-existent MX for destination domain

Hello, I have a question about how Postfix generate warning messages about non-existent domains, non-existent MX domain names or for timeout connecting to a destination mail server. In fact, my real problem is that postfix does not warn me about nothing when I incorrectly write the domain

Re: postfix3 with opendkim

On Tue, 10 Mar 2020 at 14:08, SysAdmin EM wrote: > > grep milter /etc/postfix/master.cf > # -o milter_macro_daemon_name=ORIGINATING > # -o milter_macro_daemon_name=ORIGINATING > > I can't find the problem flaw > > Any ideas? postfix 3 is compatible? > > El mar., 10 de mar. de 2020 a la(s)

Re: postfix3 with opendkim

grep milter /etc/postfix/master.cf # -o milter_macro_daemon_name=ORIGINATING # -o milter_macro_daemon_name=ORIGINATING I can't find the problem flaw Any ideas? postfix 3 is compatible? El mar., 10 de mar. de 2020 a la(s) 10:58, Dominic Raferd ( domi...@timedicer.co.uk) escribió: > On Tue, 10

Re: postfix3 with opendkim

On Tue, 10 Mar 2020 at 13:52, SysAdmin EM wrote: > > Hello.! > > In the mail.cf i add this options: > # OpenDKIM > smtpd_milters = inet:127.0.0.1:8891 > non_smtpd_milters = $smtpd_milters > milter_default_action = accept > milter_protocol = 6 > > when i send and email

Re: t/s outbound 99% timeout on TBird?

li...@sbt.net.au: > Mar 10 00:10:50 geko postfix/cleanup[9389]: C099F42B0143: > message-id=<5262b5f4-18d5-b7fb-b09a-be37f1d61b9e@a_tld.com> > Mar 10 00:12:38 geko opendkim[1322]: C099F42B0143: DKIM-Signature field > added (s=default, d=a_tld.com) You need to find out why this took TWO MINUTES.

Re: postfix3 with opendkim

Hello.! In the mail.cf i add this options: # OpenDKIM smtpd_milters = inet:127.0.0.1:8891 non_smtpd_milters = $smtpd_milters milter_default_action = accept milter_protocol = 6 when i send and email from console, i not see communication with postfix and opendkim Mar 10

RE: postfix3 with opendkim

Hi, may I ask what your Postfix config looks like for OpenDkim ? In Postfix 2.x it is close to this : smtpd_milters = inet:127.0.0.1:8891 non_smtpd_milters = $smtpd_milters milter_default_action = accept milter_protocol = 6 -ANGELO FAZZINA ang...@uconn.edu University of Connecticut, ITS,

postfix3 with opendkim

Hello, my again. I update my Postfix 2 to Postfix 3. Postfix not communicating with opendkim. Mar 10 10:14:31 server003 opendkim[18596]: OpenDKIM Filter: mi_stop=1 Mar 10 10:14:31 server003 opendkim[18596]: OpenDKIM Filter v2.11.0 terminating with status 0, errno = 0 Mar 10 10:14:34 server

Re: t/s outbound 99% timeout on TBird?

On Tue, March 10, 2020 12:33 pm, Viktor Dukhovni wrote: > > One interesting tidbit however is the 111s "before active" time in the > delays= times. > > http://www.postfix.org/postconf.5.html#delay_logging_resolution_limit > > > This does seem to suggest that it took ~111 seconds for the message

Re: postfix and systemctl

Hi! In the file /etc/yum.repos.d/gf.repo add this config: cat /etc/yum.repos.d/gf.repo [gf] name=Ghettoforge packages that won't overwrite core distro packages. mirrorlist=http://mirrorlist.ghettoforge.org/el/7/gf/$basearch/mirrorlist enabled=1 gpgcheck=1

Re: postfix and systemctl

On 10/03/20 11:50 pm, Emanuel Postfix wrote: I install postfix3 in Centos 7, I'm used to working with the systemctl command but I can't administer the service that way. Does anyone have an example of the configuration to work via systemctl? Where did you get postfix 3 from? The packages at

Re: t/s outbound 99% timeout on TBird?

On Tue, March 10, 2020 10:27 am, Wietse Venema wrote: > This is Postfix logging while SENDING email through an after-queue > content filter (which has serious congestion, but that is not the problem > in $SUBJECT). > > To come back to $SUBJECT, if you have user clients timing out, then > you

Re: postfix and systemctl

> I install postfix3 in Centos 7, I'm used to working with the systemctl > command but I can't administer the service that way. Does anyone have an > example of the configuration to work via systemctl? You should check if there is a postfix service file for systemd: [root@server ~]# rpm -ql

postfix and systemctl

Hello, I install postfix3 in Centos 7, I'm used to working with the systemctl command but I can't administer the service that way. Does anyone have an example of the configuration to work via systemctl? Thank you. Regards,