Re: Error message for a couple of mails : lost connection ...

[lst_hoe02]

Zitat von "Barchfeld, Andreas" <andreas.barchf...@kinderkrankenhaus.net>:

Hello,

for around 5% of our outgoing mails we get the message:


All mails go through our mail provider.




# cat /proc/sys/net/ipv4/tcp_window_scaling


# cat /proc/sys/net/ipv4/tcp_timestamps


# cat /proc/sys/net/ipv4/tcp_sack
1

MTU = 1500

# cat /proc/sys/net/core/[rw]mem_max
256960
256960

# cat /proc/sys/net/core/[rw]mem_default
256960
256960

# postconf -n
alias_database = hash:/etc/aliases
alias_maps = hash:/etc/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
debug_peer_level = 2
home_mailbox = Maildir/
html_directory = no
inet_interfaces = all
local_destination_concurrency_limit = 5
local_destination_recipient_limit = 200
mail_owner = postfix
mailbox_command = /usr/bin/procmail -a "$USER"
mailq_path = /usr/bin/mailq.postfix
manpage_directory = /usr/share/man
message_size_limit = 35389440

mydomain = akk.local
myhostname = postfix.akk.local
myorigin = $mydomain
newaliases_path = /usr/bin/newaliases.postfix
queue_directory = /var/spool/postfix
readme_directory = /usr/share/doc/postfix/README_FILES
relayhost = smtp.1und1.de
sample_directory = /usr/share/doc/postfix/samples
sendmail_path = /usr/sbin/sendmail.postfix
setgid_group = postdrop
smtp_data_xfer_timeout = 600
smtp_generic_maps = hash:/etc/postfix/generic
smtp_sasl_auth_enable = yes
smtp_sasl_password_maps = hash:/etc/postfix/sasl_pwds
smtp_sasl_security_options = noanonymous
unknown_local_recipient_reject_code = 550


Firewall: FortiGate 110C

"lost connection with <<mail-provider>> while sending end of data -- message may be sent more than once" There are no dependencies on the size of the mail or the mail address. It takes a few hours, then the mails are sent. Unfortunately, from these mails, some mails are send two or three times - as the message says. mydestination = $myhostname, localhost.$mydomain, localhost, $mydomain, kinderkrankenhaus.net

- Stateful firewall with SMTP "fixup" issues at your side or the remote
- Network errors regarding SACK/PMTU/Windows-Scaling/ECN whatever



Regards

Andreas

To debug the problem you need a complete tcpdump from both ends of a failing mail. What might be the problem - Content Filter (pre-queue) at the provider side which has load/concurrency issues You might try to dump-down Postfix smtp with "smtp_discard_ehlo_keywords" to not use features like ESMTP pipelining first, but as said to debug the problem tcpdumps will be preferred.

smime.p7s
Description: S/MIME Cryptographic Signature