Lost connection

[jason hirsh]

I am having trouble sending email to a specific server



I got the following error


"lost connection with
   mx.example.org [xx.xx.xx.xxx] while receiving 
the initial server
   greeting”


The operator says its my issue yet i have no problems with any other servers

my postconf -n is as follows:


body_checks = regexp:/usr/local/etc/postfix/body_check
bounce_size_limit = 5
command_directory = /usr/local/sbin
config_directory = /usr/local/etc/postfix
content_filter = smtp-amavis:[127.0.0.1]:10024
daemon_directory = /usr/local/libexec/postfix
daemon_timeout = 36000s
data_directory = /var/db/postfix
debugger_command = PATH=/bin:/usr/bin:/usr/local/bin:/usr/X11R6/bin xxgdb 
$daemon_directory/$process_name $process_id & sleep 5
delay_warning_time = 2h
disable_vrfy_command = yes
header_checks = regexp:/usr/local/etc/postfix/header_checks
home_mailbox = Maildir/
html_directory = /usr/local/share/doc/postfix
inet_protocols = ipv4
mail_owner = postfix
mail_spool_directory = /var/mail/vmail
mailq_path = /usr/local/bin/mailq
manpage_directory = /usr/local/man
maps_rbl_domains = bl.spamcop.net 
mydestination = localhost.$mydomain, localhost
mydomain = theoceanwindow-bv.com 
mynetworks = removed
myorigin = $mydomain
newaliases_path = /usr/local/bin/newaliases
queue_directory = /var/spool/postfix
readme_directory = /usr/local/share/doc/postfix
recipient_bcc_maps = hash:/usr/local/etc/postfix/recipient_bcc
relay_recipient_maps = hash:/usr/local/etc/postfix/relay_recipients
sample_directory = /usr/local/etc/postfix
sendmail_path = /usr/local/sbin/sendmail
setgid_group = maildrop
smtp_tls_note_starttls_offer = yes
smtpd_banner = $myhostname ESMTP
smtpd_delay_reject = yes
smtpd_helo_required = yes
smtpd_helo_restrictions = permit_sasl_authenticated,check_helo_access 
hash:/usr/local/etc/postfix/helo_access,reject_invalid_hostname,permit
smtpd_recipient_restrictions = permit_mynetworks, check_sender_access 
pcre://usr/local/etc/postfix/sender_access 
 reject_rhsbl_sender 
fresh.spameatingmonkey.net , 
reject_unauth_destination, check_client_access 
hash:/usr/local/etc/postfix/rbl_override, reject_rbl_client zen.spamhaus.org 
, reject_rbl_client bl.spam, reject_rbl_client 
bl.spameatingmonkey.net , reject_rhsbl_client 
fresh.spameatingmonkey.net , 
reject_rhsbl_sender urired.spameatingmonkey.net 
,permit_sasl_authenticated
smtpd_relay_restrictions = permit_mynetworks permit_sasl_authenticated
smtpd_sasl_auth_enable = yes
smtpd_sasl_path = private/auth
smtpd_sasl_type = dovecot
smtpd_sender_restrictions = permit_sasl_authenticated, 
reject_unknown_sender_domain, reject_unauthenticated_sender_login_mismatch, 
check_sender_access pcre://usr/local/etc/postfix/sender_access 
 permit_mynetworks
smtpd_tls_CAfile = /usr/local/etc/keys/root.crt
smtpd_tls_cert_file = /usr/local/etc/keys/server.cert
smtpd_tls_key_file = /usr/local/etc/keys/private.key
smtpd_tls_loglevel = 3
smtpd_tls_received_header = yes
smtpd_tls_security_level = may
smtpd_tls_session_cache_timeout = 3600s
tls_random_source = dev:/dev/urandom
unknown_local_recipient_reject_code = 550
virtual_alias_maps = hash:/usr/local/etc/postfix/virtual
virtual_gid_maps = static:1000
virtual_mailbox_base = /var/mail/vmail
virtual_mailbox_domains = hash:/usr/local/etc/postfix/virtual_domains
virtual_mailbox_maps = hash:/usr/local/etc/postfix/virtual_mailbox
virtual_minimum_uid = 100
virtual_uid_maps = static:1003

Re: Lost connection

[li...@rhsoft.net]

Am 18.10.2014 um 15:36 schrieb jason hirsh:

I am having trouble sending email to a specific server

I got the following error

"lost connection with
mx.example.org [xx.xx.xx.xxx] while receiving
the initial servergreeting”

The operator says its my issue yet i have no problems with any other servers


that is nonsense because he can't know that - no other complaints don't 
mean it happens here and there without take notice not does that message 
alone means anything at all


"lost connection" is just the messenger that, well, the connection was 
lost and is completly outside postfix scope, the only relevant question 
is how often that happens and if a lter retry works which is expected 
behavior


look below 8 lost connections to one server of hotmail

so what - 12 successful - the first was retried seconds later to a 
different MX, the reason could have been our link, their link, one of 
the routers between the servers at that moment - who cares?


so did your mail *really* fail or do you just have panic because you see 
"lost connection" for no reason?

_

cat maillog | grep "65.54.188.72" | grep "status=sent" | wc -l
12

Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, 
delays=22/121/1.5/0.84, dsn=2.0.0, status=sent (250 
 Queued mail for 
delivery)
Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, 
delays=22/121/1.5/0.84, dsn=2.0.0, status=sent (250 
 Queued mail for 
delivery)

_

Oct 12 18:00:36 mail postfix/smtp[17729]: 3jG73B02HDz2d: lost connection 
with mx2.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 12 18:08:00 mail postfix/smtp[17730]: 3jG74F4kj8z3H: lost connection 
with mx2.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 12 18:14:22 mail postfix/smtp[17729]: 3jG74q691qz3W: lost connection 
with mx3.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 12 18:23:39 mail postfix/smtp[17731]: 3jG76B1BPpz3y: lost connection 
with mx1.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 12 18:26:48 mail postfix/smtp[17728]: 3jG76L4Dkjz3s: lost connection 
with mx4.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 12 18:28:03 mail postfix/smtp[17729]: 3jG77G3z6fz45: lost connection 
with mx1.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 14 07:55:09 mail postfix/smtp[29058]: 3jH5ZM0zRgz23: lost connection 
with mx2.hotmail.com[65.54.188.72] while receiving the initial server 
greeting
Oct 15 15:07:12 mail postfix/smtp[23519]: 3jHv691gnWz2W: lost connection 
with mx4.hotmail.com[65.54.188.72] while receiving the initial server 
greeting

Re: Lost connection

[jason hirsh]

I have about 8 of these over two days all continue to be retired

I am trying to be pro-active  and was just looking if ether was something i may 
have hosed on my side

although this is the only server I have been having problems 

I really think the issue is with that server but was just looking for expert 
input



> On Oct 18, 2014, at 9:54 AM, li...@rhsoft.net wrote:
> 
> 
> Am 18.10.2014 um 15:36 schrieb jason hirsh:
>> I am having trouble sending email to a specific server
>> 
>> I got the following error
>> 
>> "lost connection with
>> mx.example.org [xx.xx.xx.xxx] while receiving
>> the initial servergreeting”
>> 
>> The operator says its my issue yet i have no problems with any other servers
> 
> that is nonsense because he can't know that - no other complaints don't mean 
> it happens here and there without take notice not does that message alone 
> means anything at all
> 
> "lost connection" is just the messenger that, well, the connection was lost 
> and is completly outside postfix scope, the only relevant question is how 
> often that happens and if a lter retry works which is expected behavior
> 
> look below 8 lost connections to one server of hotmail
> 
> so what - 12 successful - the first was retried seconds later to a different 
> MX, the reason could have been our link, their link, one of the routers 
> between the servers at that moment - who cares?
> 
> so did your mail *really* fail or do you just have panic because you see 
> "lost connection" for no reason?
> _
> 
> cat maillog | grep "65.54.188.72" | grep "status=sent" | wc -l
> 12
> 
> Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
> relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, delays=22/121/1.5/0.84, 
> dsn=2.0.0, status=sent (250 
>  Queued mail for 
> delivery)
> Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
> relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, delays=22/121/1.5/0.84, 
> dsn=2.0.0, status=sent (250 
>  Queued mail for 
> delivery)
> _
> 
> Oct 12 18:00:36 mail postfix/smtp[17729]: 3jG73B02HDz2d: lost connection with 
> mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 12 18:08:00 mail postfix/smtp[17730]: 3jG74F4kj8z3H: lost connection with 
> mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 12 18:14:22 mail postfix/smtp[17729]: 3jG74q691qz3W: lost connection with 
> mx3.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 12 18:23:39 mail postfix/smtp[17731]: 3jG76B1BPpz3y: lost connection with 
> mx1.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 12 18:26:48 mail postfix/smtp[17728]: 3jG76L4Dkjz3s: lost connection with 
> mx4.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 12 18:28:03 mail postfix/smtp[17729]: 3jG77G3z6fz45: lost connection with 
> mx1.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 14 07:55:09 mail postfix/smtp[29058]: 3jH5ZM0zRgz23: lost connection with 
> mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
> Oct 15 15:07:12 mail postfix/smtp[23519]: 3jHv691gnWz2W: lost connection with 
> mx4.hotmail.com[65.54.188.72] while receiving the initial server greeting

Re: Lost connection

[li...@rhsoft.net]

Am 18.10.2014 um 16:01 schrieb jason hirsh:

I have about 8 of these over two days all continue to be retired
I am trying to be pro-active  and was just looking if ether was something i may 
have hosed on my side
although this is the only server I have been having problems
I really think the issue is with that server but was just looking for expert 
input


try "net.ipv4.tcp_window_scaling = 0" in sysctl.conf and "sysctl -p", 
maybe they have some crap device in front of their server!


https://www.google.at/#q=smtp+tcp+window+scaling+problems


On Oct 18, 2014, at 9:54 AM, li...@rhsoft.net wrote:

Am 18.10.2014 um 15:36 schrieb jason hirsh:

I am having trouble sending email to a specific server

I got the following error

"lost connection with
mx.example.org [xx.xx.xx.xxx] while receiving
the initial servergreeting”

The operator says its my issue yet i have no problems with any other servers


that is nonsense because he can't know that - no other complaints don't mean it 
happens here and there without take notice not does that message alone means 
anything at all

"lost connection" is just the messenger that, well, the connection was lost and 
is completly outside postfix scope, the only relevant question is how often that happens 
and if a lter retry works which is expected behavior

look below 8 lost connections to one server of hotmail

so what - 12 successful - the first was retried seconds later to a different 
MX, the reason could have been our link, their link, one of the routers between 
the servers at that moment - who cares?

so did your mail *really* fail or do you just have panic because you see "lost 
connection" for no reason?
_

cat maillog | grep "65.54.188.72" | grep "status=sent" | wc -l
12

Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, delays=22/121/1.5/0.84, dsn=2.0.0, 
status=sent (250  Queued mail 
for delivery)
Oct 12 18:00:39 mail postfix/smtp[17729]: 3jG73B02HDz2d: to=<***>, 
relay=mx1.hotmail.com[65.55.37.104]:25, delay=146, delays=22/121/1.5/0.84, dsn=2.0.0, 
status=sent (250  Queued mail 
for delivery)
_

Oct 12 18:00:36 mail postfix/smtp[17729]: 3jG73B02HDz2d: lost connection with 
mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 12 18:08:00 mail postfix/smtp[17730]: 3jG74F4kj8z3H: lost connection with 
mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 12 18:14:22 mail postfix/smtp[17729]: 3jG74q691qz3W: lost connection with 
mx3.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 12 18:23:39 mail postfix/smtp[17731]: 3jG76B1BPpz3y: lost connection with 
mx1.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 12 18:26:48 mail postfix/smtp[17728]: 3jG76L4Dkjz3s: lost connection with 
mx4.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 12 18:28:03 mail postfix/smtp[17729]: 3jG77G3z6fz45: lost connection with 
mx1.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 14 07:55:09 mail postfix/smtp[29058]: 3jH5ZM0zRgz23: lost connection with 
mx2.hotmail.com[65.54.188.72] while receiving the initial server greeting
Oct 15 15:07:12 mail postfix/smtp[23519]: 3jHv691gnWz2W: lost connection with 
mx4.hotmail.com[65.54.188.72] while receiving the initial server greeting

Re: Lost connection

[Wietse Venema]

jason hirsh:
> I am having trouble sending email to a specific server
> I got the following error
> 
> "lost connection with
>mx.example.org [xx.xx.xx.xxx] while receiving 
> the initial server
>greeting?
> 
> The operator says its my issue yet i have no problems with any other servers

Try:

$ telnet xx.xx.xx.xxx 25

and report what happens.

If the host replies with "220 servername" then send:

EHLO your.client.name

and report what happens.

Wietse

Re: Lost connection

[Stephen Satchell]

On 10/18/2014 07:01 AM, jason hirsh wrote:
> I have about 8 of these over two days all continue to be retired
> 
> I am trying to be pro-active  and was just looking if ether was something i 
> may have hosed on my side
> 
> although this is the only server I have been having problems 
> 
> I really think the issue is with that server but was just looking for expert 
> input

What happens when you use TELNET to connect to the remote server on port
25?  Perform the test repeatedly from your mail server -- that will help
you with clue.  For example, if it takes a long time for TELNET to
connect, then you know the remote server may be overloaded.

I discovered with, a TACACS daemon with a too-short backlog, that the IP
stack in the server will complete the three-part handshake -- then finds
it can't pass the connection to the LISTEN socket because the backlog is
full.  After the various TCP wait times expire, the stack sends RST and
forgets the connection.  The key for making me think this is the
scenario is the clause "while receiving the initial servergreeting" in
the log message.

The remote SMTP server may never have had the connection passed on to it
because of high backlog overflow on the remote host.

One reason the remote operator doesn't see a problem is because s/he
never knows this is happening (no logging of the event) and no other
mail server operator has complained to him about this.  I know I would
ignore the message when I see it.

Are you getting bounce messages on mail being sent to this server?  Are
you seeing this happen again and again on the same queued message, or it
is happening with different queued mail?

Re: Lost connection

[jason hirsh]

I get this


Trying 66.96.142.51...
Connected to 51.142.96.66.static.eigbox.net.
Escape character is '^]'.
220 bosimpinc11 bizsmtp ESMTP server ready




I am getting the original; error message on all mail i end only tot that server



> On Oct 18, 2014, at 10:21 AM, Stephen Satchell  wrote:
> 
> On 10/18/2014 07:01 AM, jason hirsh wrote:
>> I have about 8 of these over two days all continue to be retired
>> 
>> I am trying to be pro-active  and was just looking if ether was something i 
>> may have hosed on my side
>> 
>> although this is the only server I have been having problems 
>> 
>> I really think the issue is with that server but was just looking for expert 
>> input
> 
> What happens when you use TELNET to connect to the remote server on port
> 25?  Perform the test repeatedly from your mail server -- that will help
> you with clue.  For example, if it takes a long time for TELNET to
> connect, then you know the remote server may be overloaded.
> 
> I discovered with, a TACACS daemon with a too-short backlog, that the IP
> stack in the server will complete the three-part handshake -- then finds
> it can't pass the connection to the LISTEN socket because the backlog is
> full.  After the various TCP wait times expire, the stack sends RST and
> forgets the connection.  The key for making me think this is the
> scenario is the clause "while receiving the initial servergreeting" in
> the log message.
> 
> The remote SMTP server may never have had the connection passed on to it
> because of high backlog overflow on the remote host.
> 
> One reason the remote operator doesn't see a problem is because s/he
> never knows this is happening (no logging of the event) and no other
> mail server operator has complained to him about this.  I know I would
> ignore the message when I see it.
> 
> Are you getting bounce messages on mail being sent to this server?  Are
> you seeing this happen again and again on the same queued message, or it
> is happening with different queued mail?
> 
> 

Re: Lost connection

[Wietse Venema]

jason hirsh:
> I get this
>
> Trying 66.96.142.51...
> Connected to 51.142.96.66.static.eigbox.net.
> Escape character is '^]'.
> 220 bosimpinc11 bizsmtp ESMTP server ready

Then try step 2 in my reply:

EHLO your.server.name

and report what happens.

I see the following:

% telnet 66.96.142.51 25
Trying 66.96.142.51...
Connected to 51.142.96.66.static.eigbox.net.
Escape character is '^]'.
220 bosimpinc11 bizsmtp ESMTP server ready
EHLO your.server.name
250-bosimpinc11 hello [70.104.130.26], pleased to meet you
250-HELP
250-SIZE 3000
250-8BITMIME
250-STARTTLS
250 OK
quit
221 bosimpinc11 bizsmtp closing connection
Connection closed by foreign host.

Wietse

Re: Lost connection

[jason hirsh]

> On Oct 18, 2014, at 10:30 AM, jason hirsh  wrote:
> 
> I get this
> 
> 
> Trying 66.96.142.51...
> Connected to 51.142.96.66.static.eigbox.net .
> Escape character is '^]'.
> 220 bosimpinc11 bizsmtp ESMTP server ready
> 
> 
> 
> 
> I am getting the original; error message on all mail i end only tot that 
> server


correcting typos should use my glasses

I get the error message that i have in the beginning of this thread on all 
traffic i send to this mail server from mine
> 
> 
> 
>> On Oct 18, 2014, at 10:21 AM, Stephen Satchell > > wrote:
>> 
>> On 10/18/2014 07:01 AM, jason hirsh wrote:
>>> I have about 8 of these over two days all continue to be retired
>>> 
>>> I am trying to be pro-active  and was just looking if ether was something i 
>>> may have hosed on my side
>>> 
>>> although this is the only server I have been having problems 
>>> 
>>> I really think the issue is with that server but was just looking for 
>>> expert input
>> 
>> What happens when you use TELNET to connect to the remote server on port
>> 25?  Perform the test repeatedly from your mail server -- that will help
>> you with clue.  For example, if it takes a long time for TELNET to
>> connect, then you know the remote server may be overloaded.
>> 
>> I discovered with, a TACACS daemon with a too-short backlog, that the IP
>> stack in the server will complete the three-part handshake -- then finds
>> it can't pass the connection to the LISTEN socket because the backlog is
>> full.  After the various TCP wait times expire, the stack sends RST and
>> forgets the connection.  The key for making me think this is the
>> scenario is the clause "while receiving the initial servergreeting" in
>> the log message.
>> 
>> The remote SMTP server may never have had the connection passed on to it
>> because of high backlog overflow on the remote host.
>> 
>> One reason the remote operator doesn't see a problem is because s/he
>> never knows this is happening (no logging of the event) and no other
>> mail server operator has complained to him about this.  I know I would
>> ignore the message when I see it.
>> 
>> Are you getting bounce messages on mail being sent to this server?  Are
>> you seeing this happen again and again on the same queued message, or it
>> is happening with different queued mail?
>> 
>> 
> 

Re: Lost connection

[jason hirsh]

oops missed step 2


I get this


Trying 66.96.142.51...
Connected to 51.142.96.66.static.eigbox.net.
Escape character is '^]'.
220 bosimpinc11 bizsmtp ESMTP server ready
EHLO mail.kasdivi.com
250-bosimpinc11 hello [209.160.65.133], pleased to meet you
250-HELP
250-SIZE 3000
250-8BITMIME
250-STARTTLS
250 OK

> On Oct 18, 2014, at 10:38 AM, Wietse Venema  wrote:
> 
> jason hirsh:
>> I get this
>> 
>> Trying 66.96.142.51...
>> Connected to 51.142.96.66.static.eigbox.net.
>> Escape character is '^]'.
>> 220 bosimpinc11 bizsmtp ESMTP server ready
> 
> Then try step 2 in my reply:
> 
> EHLO your.server.name
> 
> and report what happens.
> 
> I see the following:
> 
> % telnet 66.96.142.51 25
> Trying 66.96.142.51...
> Connected to 51.142.96.66.static.eigbox.net.
> Escape character is '^]'.
> 220 bosimpinc11 bizsmtp ESMTP server ready
> EHLO your.server.name
> 250-bosimpinc11 hello [70.104.130.26], pleased to meet you
> 250-HELP
> 250-SIZE 3000
> 250-8BITMIME
> 250-STARTTLS
> 250 OK
> quit
> 221 bosimpinc11 bizsmtp closing connection
> Connection closed by foreign host.
> 
>   Wietse

Re: Lost connection

[Wietse Venema]

jason hirsh:
> oops missed step 2
> 
> 
> I get this
> 
> 
> Trying 66.96.142.51...
> Connected to 51.142.96.66.static.eigbox.net.
> Escape character is '^]'.
> 220 bosimpinc11 bizsmtp ESMTP server ready
> EHLO mail.kasdivi.com
> 250-bosimpinc11 hello [209.160.65.133], pleased to meet you
> 250-HELP
> 250-SIZE 3000
> 250-8BITMIME
> 250-STARTTLS
> 250 OK

That is interesting. These are the same commands that Postfix would
send.

Are you making the telnet connection from tyhe machine that runs
Postfix? (mail.kasdivi.com is 209.160.65.133, so it looks like you are)

Do you have any "-o name=value" settings in master.cf? If running
Postix 2.11, try

$ postconf -P '*/inet/*'

Otherwise, you need to look at master.cf yourself.

Wietse

Re: Lost connection

[Stephen Satchell]

On 10/18/2014 07:01 AM, jason hirsh wrote:
> I have about 8 of these over two days all continue to be retired
> 
> I am trying to be pro-active  and was just looking if ether was something i 
> may have hosed on my side

I forgot to ask:  what does your DNS entries look like for your mail
server?  Specifically:

  A record
  PTR record
  MX record(s)

Back when I was working for a web hosting company, I insisted that any
incoming mail come from a launch point with best practices DNS.  My
policy filter would return a polite error message and close the
connection; this guy might just be rude.

My tests:
  1.  IP address has a PTR record with a FQDN that looks to be
statically assigned.  This was developed over time, as I learned the
patterns. I also has a whitelist of REGEXP patterns.  Multiple returns
were a no-no.

  2.  Look-up on FQDN returns an A record with the same IP address.
Multiple IP addresses can be returned, but one of them must match the IP
address of the incoming connection

  3.  Look-up of the domain name (tried several variations) returns MX
record(s).  I allowed for those large groups who split incoming mail
from outgoing mail in a server farm -- the idea here is that I didn't
accept mail from any endpoint that may not have a postmaster associated
with it.

Re: Lost connection

[jason hirsh]

> On Oct 18, 2014, at 10:54 AM, Wietse Venema  wrote:
> 
> jason hirsh:
>> oops missed step 2
>> 
>> 
>> I get this
>> 
>> 
>> Trying 66.96.142.51...
>> Connected to 51.142.96.66.static.eigbox.net.
>> Escape character is '^]'.
>> 220 bosimpinc11 bizsmtp ESMTP server ready
>> EHLO mail.kasdivi.com
>> 250-bosimpinc11 hello [209.160.65.133], pleased to meet you
>> 250-HELP
>> 250-SIZE 3000
>> 250-8BITMIME
>> 250-STARTTLS
>> 250 OK
> 
> That is interesting. These are the same commands that Postfix would
> send.
> 
> Are you making the telnet connection from tyhe machine that runs
> Postfix? (mail.kasdivi.com  is 209.160.65.133, so 
> it looks like you are)

Yes I am running Postfix version 2.11-20131001
> 
> Do you have any "-o name=value" settings in master.cf?

not that i can see

> If running
> Postix 2.11, try
> 
>$ postconf -P '*/inet/*’

I get a "postconf: illegal option — P"
> 
> 
> Otherwise, you need to look at master.cf yourself.
> 
>   Wietse

Re: Lost connection

[Wietse Venema]

jason hirsh:
> 
> > On Oct 18, 2014, at 10:54 AM, Wietse Venema  wrote:
> > 
> > jason hirsh:
> >> oops missed step 2
> >> 
> >> 
> >> I get this
> >> 
> >> 
> >> Trying 66.96.142.51...
> >> Connected to 51.142.96.66.static.eigbox.net.
> >> Escape character is '^]'.
> >> 220 bosimpinc11 bizsmtp ESMTP server ready
> >> EHLO mail.kasdivi.com
> >> 250-bosimpinc11 hello [209.160.65.133], pleased to meet you
> >> 250-HELP
> >> 250-SIZE 3000
> >> 250-8BITMIME
> >> 250-STARTTLS
> >> 250 OK
> > 
> > That is interesting. These are the same commands that Postfix would
> > send.

Does the "lost connection" happen because Postfix makes multiple
connections at the same time? Try this example:

/etc/postfix/main.cf:
transport_maps = hash:/etc/postfix/transport
slow_destination_concurrency_limit = 1
slow_initial_destination_concurrency = 1
slow_destination_concurrency_failed_cohort_limit = 10

/etc/postfix/transport:
example.com  slow:

/etc/postfix/master.cf:
# service type  private unpriv  chroot  wakeup  maxproc command
slow  unix -   -   n   -   -smtp
-o smtp_connection_cache_on_demand=no

for "example.com" specify the problem domain name (not hostname
or IP address).

Wietse

Re: Lost connection

[jason hirsh]

> On Oct 18, 2014, at 11:04 AM, Richard  
> wrote:
> 
> 
> 
>  Original Message 
>> Date: Saturday, October 18, 2014 10:45:09 -0400
>> From: jason hirsh mailto:hir...@att.net>>
>> To: Postfix users > >
>> Cc: 
>> Subject: Re: Lost connection
>> 
>> oops missed step 2
>> 
>> 
>> I get this
>> 
>> 
>> Trying 66.96.142.51...
>> Connected to 51.142.96.66.static.eigbox.net.
>> Escape character is '^]'.
>> 220 bosimpinc11 bizsmtp ESMTP server ready
>> EHLO mail.kasdivi.com
>> 250-bosimpinc11 hello [209.160.65.133], pleased to meet you
>> 250-HELP
>> 250-SIZE 3000
>> 250-8BITMIME
>> 250-STARTTLS
>> 250 OK
> 
> 
> 
> If "mail.kasdivi.com " is your mail server's true 
> name, then the
> forward and reverse lookups don't match. There is an rDNS record for
> 209.160.65.133, but it isn't "mail.kasdivi.com ". 
> Some mail servers
> will refuse mail unless the forward and reverse match. What they
> tell you may vary. 
> 
> Unless/until these match, I think that the onus falls to your side.
> 



In my master record I do show this PTR for the domain in question

I have 7 domains on this server

tuna.theoceanwindoiw-bv.com  is the server 

> mail.kasdivi.com .  38400   IN  A   
> 209.160.65.133
> 
> 133.65.160.209.in-addr.arpa. 86188 IN PTR tuna.theoceanwindow-bv.com 
> .
> 
> 
> 
>   - Richard

Re: Lost connection

[jason hirsh]

I made these changes and my mail log indicates that the delays in delivery have 
been reduced 
with no loss of connection 
I presume that test mail has been delivered, at least it was delivered to the 
other mail server










> On Oct 18, 2014, at 11:11 AM, Wietse Venema  wrote:
> 
> jason hirsh:
>> 
>>> On Oct 18, 2014, at 10:54 AM, Wietse Venema  wrote:
>>> 
>>> jason hirsh:
 oops missed step 2
 
 
 I get this
 
 
 Trying 66.96.142.51...
 Connected to 51.142.96.66.static.eigbox.net.
 Escape character is '^]'.
 220 bosimpinc11 bizsmtp ESMTP server ready
 EHLO mail.kasdivi.com
 250-bosimpinc11 hello [209.160.65.133], pleased to meet you
 250-HELP
 250-SIZE 3000
 250-8BITMIME
 250-STARTTLS
 250 OK
>>> 
>>> That is interesting. These are the same commands that Postfix would
>>> send.
> 
> Does the "lost connection" happen because Postfix makes multiple
> connections at the same time? Try this example:
> 
> /etc/postfix/main.cf:
>transport_maps = hash:/etc/postfix/transport
>slow_destination_concurrency_limit = 1
>slow_initial_destination_concurrency = 1
>slow_destination_concurrency_failed_cohort_limit = 10
> 
> /etc/postfix/transport:
>example.com   slow:
> 
> /etc/postfix/master.cf:
># service type  private unpriv  chroot  wakeup  maxproc command
>slow  unix -   -   n   -   -smtp
>-o smtp_connection_cache_on_demand=no
> 
> for "example.com " specify the problem domain name (not 
> hostname
> or IP address).
> 
>   Wietse

Re: DROP not a Postfix problem!!: OT: can anybody tell me what is wrong here

[John]

*Thank you.*
Yes that is/was the problem!
An automatic update to my system broke DNSSEC tools, so it did not roll, 
no roll, no notice to update the DLV 

Temp solution, drop DNSSEC while I try to fix the DNSSEC tools.
On 17/10/2014 6:55 AM, wie...@porcupine.org (Wietse Venema) wrote:

John:

On 14/10/2014 11:14 PM, Claus Assmann wrote:

On Tue, Oct 14, 2014, John wrote:

 host MX.cogeco.ca[216.221.81.26] said: 451 Postmaster
   Code 5 - #4.1.8 Domain of sender address   does not 
resolve.

Try again, it's just a temporary error.

BTW: cool error message with those 8 bit chars in it:
utilis? pour envoyer le courriel, ne se r?sout pas

Some REMOTE MAIL SERVER is not accepting your mail.  If you want
to know why, ask the people who are responsible for the REMOTE MAIL
SERVER.

Wietse

Re: Lost connection

[Viktor Dukhovni]

On Sat, Oct 18, 2014 at 11:01:29AM -0400, jason hirsh wrote:

> > Are you making the telnet connection from tyhe machine that runs
> > Postfix? (mail.kasdivi.com  is 209.160.65.133, so 
> > it looks like you are)
> 
> Yes I am running Postfix version 2.11-20131001

Postfix 2.11.0 was released in January 2014, and 2.11.2 since.
You should no longer be using 2.11 snapshots.

-- 
Viktor.