Postfix - antispam and relay package

garthk

Works great but I have a question. There is a company sending us email with a single MX record, say mail.company.com, but the email is actually being sent by one of multiple servers, mail1.company.com, mail2.company.com, and so on. None of these servers has a DNS record so can not be found by PF after the RCPT TO: is received. This causes the email to be rejected, correctly IMHO, but I need to figure out how to let this email thru. Can I whitelist these servers and, if so, how?

Thanx,
Garth

garthk

Sorry to reply to my own post but… the initial HELO is from mail.company.com and is resolvable. Prob is, that's not the server that actually sends the email and those servers are not resolvable.

Thanx,
Garth

biggsy

If you can tell whether they're in the same subnet you can whitelist that subnet under Access Lists > CIDR

Like:

10.20.30.0/24 permit

Bismarck

@garthk:

Sorry to reply to my own post but… the initial HELO is from mail.company.com and is resolvable. Prob is, that's not the server that actually sends the email and those servers are not resolvable.

Thanx,
Garth

https://forum.pfsense.org/index.php?topic=40622.msg428403#msg428403

Bismarck

@sbillmann:

2. Some mails take a very long time to get delivered to my actual mail server. I guess this is because some bigger companies with multiple mail servers send mails out through a different server once the message isn't accpeted instantly by postfix. (gmail or hrs for example)
Is there a way to accept e-mails faster even if the initial sender ip differs from the current sender ip in postfix?

And again thank you (in advance)

Many apologies if this has been asked and answered before.

https://forum.pfsense.org/index.php?topic=40622.msg425790#msg425790

azekiel

Is the postscreen cache now persistent (normally it would be deleted after a restart of the service)?

If not, why not use postgrey then? This one works the same way as postscreen does and the persistent cache does work!

Greets

azekiel

another question: how to disable the recipient check? i remove the part from smtpd_recipient_restrictions but is there a way in the gui?

dene14

Pretty nice module! Thanks for your great work…

+1 for CertManager's certificate support for STARTTLS... It looks a bit strange when you have to generate SelfSigned with certmanager, download cert + key, and upload them by scp to router... Also this conf doesn't survives reinstalls from backup :(

However it seems I've found a bug in current version:
when I select "listen on all Interfaces/IPs"
this line appears in main.cf. unfortunately, it isn't valid
"inet_interfaces = "

to fix that we need to bind that selection to
"inet_interfaces = all"

Thanks!

sandroditommaso

Hi, I'm using this nice package for a few months without major problems.
Just one thing…
Every day I find some incoming emails in the "incoming" state that are not delivered .
Why?

Bismarck

Hello marcelloc, will

/usr/sbin/pkg_add -r p5-perl-ldap

still work with 2.2/10.1? If I remember right there is no pkg_add anymore with FreeBSD 10.1 and what wil happen with packages that have been installed via pkg_add on 8.3 FreeBSD and upgraded to FreeBSD 10.1, will they sill work?

Thanks for all.

marcelloc

@Bismarck:

Hello marcelloc, will

/usr/sbin/pkg_add -r p5-perl-ldap

still work with 2.2/10.1?

On freebsd 10, use pkg add instead of pkg_add

Bismarck

Thanks marcelloc, always appreciated.

https://doc.pfsense.org/index.php/Installing_FreeBSD_Packages#pfSense_2.2

hrtraveler

Having problems with Postfix Forwarder after upgrading to pfSense 2.2 (worked fine on 2.15 immediately before upgrading).

Mail no longer goes out, and client give a time out message. Testing though WebGui > Diagnostics > Test Port I can make a connection on port 25 and I don't see anything in the firewall log that would make me believe that this is a firewall issue, therefore I'm left with Postfix.

Upon restarting Postfix I get the following in the log.

Jan 24 16:54:33	postfix/postfix-script[55658]: stopping the Postfix mail system
Jan 24 16:54:33	postfix/master[81995]: terminating on signal 15
Jan 24 16:54:35	postfix/postfix-script[87134]: warning: not owned by root: /var/spool/postfix
Jan 24 16:54:35	postfix/postfix-script[91266]: starting the Postfix mail system
Jan 24 16:54:35	postfix/master[96254]: daemon started -- version 2.11.3, configuration /usr/local/etc/postfix
Jan 24 16:54:35	postfix/master[96254]: warning: process /usr/local/libexec/postfix/pickup pid 96570 exit status 1
Jan 24 16:54:35	postfix/master[96254]: warning: /usr/local/libexec/postfix/pickup: bad command startup -- throttling
Jan 24 16:54:35	postfix/master[96254]: warning: process /usr/local/libexec/postfix/qmgr pid 96714 exit status 1
Jan 24 16:54:35	postfix/master[96254]: warning: /usr/local/libexec/postfix/qmgr: bad command startup -- throttling

and then the following appears in the log on an ongoing basis (every minute or so)

Jan 24 16:57:35	postfix/master[96254]: warning: process /usr/local/libexec/postfix/pickup pid 50520 exit status 1
Jan 24 16:57:35	postfix/master[96254]: warning: /usr/local/libexec/postfix/pickup: bad command startup -- throttling
Jan 24 16:57:35	postfix/master[96254]: warning: process /usr/local/libexec/postfix/qmgr pid 50792 exit status 1
Jan 24 16:57:35	postfix/master[96254]: warning: /usr/local/libexec/postfix/qmgr: bad command startup -- throttling

Finally when I connect on port 25 is see the following

Jan 24 17:00:22	postfix/master[96254]: warning: process /usr/local/libexec/postfix/smtpd pid 94067 exit status 1
Jan 24 17:00:22	postfix/master[96254]: warning: /usr/local/libexec/postfix/smtpd: bad command startup -- throttling

I've tried reinstalling the package, though that didn't resolve the issue. Postfix is the only package I have installed.

Thanks for your help.

marcelloc

May be related to pfsense 2.2 security sysctrl option to do not allow non root users to listen on low ports.

Try to listen postfix on a high port(>1024) and nat 25 to it.(similar procedure while using carp).

It was fixed some weeks ago on squid package but it looks like something has changed or I've missed something on my tests.

hrtraveler

I moved to port 1050, but no change:

Jan 24 18:41:43	postfix/master[4057]: warning: process /usr/local/libexec/postfix/pickup pid 84681 exit status 1
Jan 24 18:41:43	postfix/master[4057]: warning: /usr/local/libexec/postfix/pickup: bad command startup -- throttling
Jan 24 18:41:43	postfix/master[4057]: warning: process /usr/local/libexec/postfix/qmgr pid 84757 exit status 1
Jan 24 18:41:43	postfix/master[4057]: warning: /usr/local/libexec/postfix/qmgr: bad command startup -- throttling
Jan 24 18:41:50	postfix/master[4057]: warning: process /usr/local/libexec/postfix/smtpd pid 84802 exit status 1
Jan 24 18:41:50	postfix/master[4057]: warning: /usr/local/libexec/postfix/smtpd: bad command startup -- throttling

Same thing if I connect directly to port 1050 or via the NAT rule from port 25.

biggsy

Same problem reported here last month - but I've just realized that I posted it in the 2.2 snapshots area.

vc6SfV8

I am also experiencing the same problem as hrtraveler after upgrading to 2.2.

marcelloc

I've updated database log integration from sqlite2 to sqlite3 but if I send a pull request for it then postfix will not work on 2.1.

So until we find a way to fix it on current pfsense 2.2 pbi, I suggest to use postfix on 2.1(as a server for exemple on virtual machine)

TeeJay

Having excactly the same problem on a brand new installation. I was looking for a substitute for the Endian Comunnity FW and pfsense certainly looks very promising, but wihout postfix it will not be very usefull to me. Is there any chance this problem will be fixed in the near future?

guyp

Ran into this today… Really need a quick and dirty fix until it can be fixed fully.

I can't roll back to the old version, as the FW is 8000 Miles away from me :(