New answers tagged

1

As defined in RFC 7208, 1.1.3, SPF is not tested against the RFC 5322 From header, but against the envelope sender i.e. the address in RFC 5321 MAIL FROM command. from=<www-data@rec15.appleandroidemail.mx> Therefore, rec15.appleandroidemail.mx is the domain of the envelope sender, and this hostname doesn't have an SPF record. You'd need an additional ...


0

I struggled with this for about an hour before figuring out via strace -f -p piding the running postfix process that it couldn't find my sasldb2 file because it was putting itself in a chroot. Edit /etc/postfix/master.cf and put an n in the chroot column. Restart postfix. Should work now.


0

Your name appears to have a dozen or more IPv6 addresses set in the DNS, but as far as I can tell none of them actually respond on any port. Mail providers which also use IPv6 (such as Google or Microsoft and many others) will try those addresses and get a timeout, then queue the mail to try again later. Your server does answer SMTP on IPv4, but it can take ...


0

Your SPF records is configured as a "Soft Fail", which allows the mail even when it fails. This is for testing purposes. You need to configure your SPF records as a "Hard Fail" by setting -all instead of ~all-


1

Postfix tuning is not necessary - start to tune Your knowledge about E-Mail System, base on Your questions You have no idea how its working. Your Postfix client (in that case is not a server) could have 512MB RAM and 1 CPU - what is important - You need fast connection Your Postfix client is not important in that case - important is E-Mail Server to which ...


1

The destination mail server (for mail.de) has chosen not to disclose the reason it rejected the mail. That reason should normally appear in the Diagnostic-Code line, but here they have replaced it with garbage. You can guess that they reject mail from freedns hosted domains, which is a reasonable thing to do. But to be 100% sure of the reason for rejection, ...


0

Minimum requirements of MTA connecting to other MTAs are: Your MTA use HELO with a correct server name. Your server name resolved into source IP address. Source IP is resolved into that name. If domain have SPF records - IP address should be allowed to send mail from this domain. Also, not required but strictly needed is not having that IP in spam ...


1

Generally: MX records tell people how to reach your email server. You need to point your MX record to wherever you want people's email servers to send email to. SPF records tell people who is allowed to send email as you. You need to include any mail servers that are allowed to send email on your behalf. DKIM records tell people what your signature looks ...


0

Have a look on this article Simple Mail Transfer Protocol: The Simple Mail Transfer Protocol (SMTP) is a communication protocol for electronic mail transmission. As an Internet standard, SMTP was first defined in 1982 by RFC 821, and updated in 2008 by RFC 5321 to Extended SMTP additions, which is the protocol variety in widespread use today. Mail servers ...


0

The ubuntu 16.04 OpenDKIM solution spawns OpenDKIM thus: /usr/sbin/opendkim -x /etc/opendkim.conf -u opendkim \ -P /var/run/opendkim/ope dkim.pid -p inet:8891@localhost The solution was to use /etc/opendkim.conf rather than /etc/opendkim/opendkim.conf. (Somewhere I'd read /etc/opendkim/opendkim.conf, correctly or incorrectly, and I forever after read ...


2

Once messages have worked through postfix there's no memory of them, meaning that the solution is the same as for re-sending messages in general, which is to use the Resent-To header. You'd need to do it with a loop, and depending on the number probably rate-limit yourself to avoid issues with Gmail. Along these lines, adjust as necessary. MDIR=/home/user/...


0

Basically SMTP require your hostname and IP being resolved properly and this requirement is checked during connect and HELO phase of any connection. Your hostname (which is delivered to other party via HELO) should have DNS A record with the address your MTA is connecting from. And this IP address should have PTR record leading to this hostname as well. ...


0

modify/change /etc/postfix/main.cf like this below: smtpd_tls_cert_file=/etc/letsencrypt/live/host.hostprovider.com/fullchain.pem to smtpd_tls_cert_file=/etc/letsencrypt/live/host.hostprovider.com/cert.pem smtpd_tls_cert_file should point to cert.pem file and not fullchain.pem


1

Restart the log server that actually writes the log entries (in your case, probably rsyslog). It was not the daemons postfix and dovecot that generate the timestamp, they just put their messages into syslog and rsyslogd proceded to format them with its configured time stamp format (and currently selected timezone translation). General solution: To know which ...


1

Start by rejecting completely invalid mail by adding reject_non_fqdn_recipient and reject_unknown_recipient_domain to appropriate restriction table. Postfix can additionally REJECT messages via check_recipient_mx_access and check_recipient_access, but how you generate an appropriate lookup depends on your acceptable level of error (rejecting non-typos / non-...


0

Lets unpack this error message lmtp(joeblow)<17980>: This is the LMTP service talking to you error: sieve: The LDA Sieve plugin does not have permission to save global Sieve script binaries; You are applying some sieve filtering globally, and the LMTP service (more specifically, the delivery part, hence the LDA term) is configured to call sieve ...


1

One way is to use the ClamAV milter. Scan outbound messages in the usual way, e.g. smtpd_milters = unix:/var/run/clamav/clamav-milter milter_default_action = accept In the milter config, set MaxFileSize 10M AddHeader Replace If the milter rejects the message, processing continues. But the X-Virus headers will be different. You can then use this header, ...


0

Make sure all packages below are installed, there are probably a package or two you might not need. This is just copied from my notes on my last install and I remember that I had the same issue and I was missing a package and I added it to my notes but I don't remember which one it was exactly. dnf -y install amavisd-new spamassassin clamav-server clamav-...


0

update: Had to ufw allow 587 and 993 to make it work...


0

On Thunderbird security.tls.version.min is 3 by default. If is an old postfix / dovecot, you have to set security.tls.version.min to 1 or to upgrade postfix , dovecot.


0

Does this issue only occur when mails receive from office 365 tenants? Has it happened before? Have you changed anything on your server? Are there any mail flow rules in office 365 which maybe cause this issue? For more details: Mail flow rule actions in Exchange Online


0

Do not set your system hostname to your naked domain name. Do not set Postfix's mailname to the naked domain name. Do not add the naked domain name to Postfix's mydestination, or to any other Postfix configuration option. Doing any of these will cause Postfix to try to deliver the mail locally rather than sending it out to the Internet. Name the system with ...


0

Solved by adding sudo apt install ca-certificates sudo postconf -e 'smtp_tls_CAfile = /etc/ssl/certs/ca-certificates.crt' sudo postfix reload Result: postfix/smtp[11023]: Trusted TLS connection established to ...


0

In DMARC the rules about which address matters are described as identifier alignment, and describe the manner in which the SPF-authenticated domain must match the domain the From-Header. You are looking at logs for a mail that is not aligned in that sense: I can tell from your opendmarc logs that you were receiving a mail that was checked for alignment with ...


0

Eventually, after much pain, I realized that the issue here had nothing to do with certbot/Letsencrypt (as stated by Michael Hampton in the comments above), but was related to Virtualmin, which has "its own way" to present certificates through Apache. So, after manually issuing certs via certbot, the solution here was to update/refresh certs via ...


0

You put your domain name in mydestination. This tells Postfix that it is meant to handle mail being sent to the domain, so it tries to deliver such mail locally. For a mail server that is meant to relay everything, the naked domain name should not be in mydestination or in the file /etc/mailname. (A subdomain is fine.) Further, the system hostname should ...


1

Actually I solved using the transport directive. I added to my main.cf the following: transport_maps = hash:/etc/postfix/transport Then I created the transport file using * smtp:[smtp-relay.gmail.com]:465 that will deliver all the email using gmail smtp (internal and external). and then I run postmap hash:/etc/postfix/transport Finally I restarted ...


0

The error message says: cannot connect to saslauthd server: No such file or directory. The two most likely explanations is either that saslauthd is not running, or that smtpd and saslauthd disagree on the name.


2

There might be a way with OpenDKIM but you can at least do it in postfix with header_checks. Create /etc/postfix/header_checks with: /^DKIM-Signature: / FILTER smtp:127.0.0.1:10025 Then in /etc/postfix/master.cf make two changes. First duplicate the SMTP listener so that it also listens on port 10025 but exclude the OpenDKIM milter. To the listen port, ...


0

If postfix is acting as the server (i.e. SMTP daemon) and will be performing the authentication, then you need to put the authorization in smtpd_. From the question it seems you've put it in smtp_ (client). Something like smtpd_sasl_security_options = noanonymous smtpd_sasl_auth_enable = yes Typically the authorization would not go in sasl_passwd as that ...


1

First, rule out issues that have little to do with the actual signing: Is your signature failing to verify, or is the recipient just failing to grab your key? (Authentication-Results: host; dkim=fail versus Authentication-Results: host; dkim=permerror)? Are you actually looking at forwarded mail (mails where you were not the party transmitting it to the ...


1

I'm using Debian buster and I have installed postfix through apt-get install. How I solved the connection time out and connection rejected issues: In the /etc/postfix/main.cf you need to have relayhost = [smtp.server.com]:PORT Instead of relayhost = smtp.server.com:PORT It took me a couple of hours to find this solution. You need the brackets. How I solved ...


0

OK, I was able to get authentication working by adding the following lines to the postfix main.cf config file: smtpd_sasl_path = private/auth smtpd_sasl_type = dovecot smtpd_sasl_auth_enable = yes


0

I found another explanation http://www.postfix.org/TUNING_README.html maximal_queue_lifetime (default: 5 days) How long a message stays in the queue before it is sent back as undeliverable. Specify 0 for mail that should be returned immediately after the first unsuccessful delivery attempt. bounce_queue_lifetime (default: 5 days, available with Postfix ...


-1

In my case was because selinux didn't allow apache (PHP) to send emails. Fixed with: setsebool -P httpd_can_sendmail=1


2

Your MX record is incorrect. $ host -t mx theomnihealthgroup.com theomnihealthgroup.com mail is handled by 1 mail. But your mail server is at mail.theomnihealthgroup.com. Edit your DNS MX record to correct this.


2

I'm not an expert on mail daemons, but think the person sending the mail was doing so from the actual mail server (client= should be the ip of the person connecting to the mail daemon). It could have been sent by a logged in user or a php script on the postfix server that uses the mail() function, or something along those lines. By default postfix and most ...


2

As usual, there is more than one way to skin the cat. But dovecot-backup exists, so have a look, it might be just the ticket for you. Some commenters advised you to check your configuration and that is good advice. In case you feel overwhelmed by where and what to check, I suggest you start with doveconf -n|less and check for the word 'location', in ...


Top 50 recent answers are included