lost connection after DATA (0 bytes)

[CXliving]

Hi Guys
I have noticed and massive drop off in mail received (notably to my spam folder that i check daily) and then getting reports of mail getting returned to senders after lengthy delays.
I have trawled your and other Postfix forums, but nothing suggested has provided any relief. Basically, I found identical issues that all seemed to be resolved by editing the MTU on either the router or NIC of the server involved, but I have checked mine and the Bayes router was 1500 and the NIC was 1492. Just in case, I changed the router to 1492, but no difference.
Despite the 256K T-Dis line this server is on, it has never battled to receive vast quantities of mail, and has seldom if EVER reached the default concurrency limit of 100 connections (I searched the last few maillog.* and found none recorded).
Then, on Sunday (5th) night I noticed the most bizarre bandwidth usage.... My line showed a consistent download of 90k and upload of roughly 20k. During the days, it obviously went up as users pulled mail and more came in, but during the nights, it never ever drops below 90k down and the little trickle up.
I was getting a few "timeout after connect" reports in maillog, but now i am getting "timeout after DATA (0 bytes)" hundreds of times and less and less legitimate mail is getting through.
If I turn off postfix, the line usage drops to zero. Give it 5 mins, turn it on and the connections come flooding in.
Then, shortly, I start seeing "timeout after DATA (0 bytes)" from loads of different hosts.
This is Postfix 2.5.6 on Centos 5 - this server has been running like a dream for ages... I have NO idea what I did that night have sparked this off, and any suggestions will be gratefully persued.
thanks so much.
cheers,
Rex

ps. postconf...
[root@mail postfix]# postconf -n
alias_maps = hash:/etc/postfix/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
daemon_directory = /usr/libexec/postfix
data_directory = /var/lib/postfix
debug_peer_level = 2
default_destination_concurrency_limit = 20
disable_vrfy_command = yes
header_checks = regexp:/etc/postfix/header_checks
html_directory = no
in_flow_delay = 1s
local_destination_concurrency_limit = 2
mail_owner = postfix
mail_spool_directory = /var/spool/mail
mailq_path = /usr/bin/mailq.postfix
manpage_directory = /usr/share/man
mydestination = $myhostname, $mydomain
mydomain = ****.co.za
myhostname = mail.****.co.za
mynetworks = *.*.*.*/27, *.*.*.*/24, 127.0.0.0/8
myorigin = $mydomain
newaliases_path = /usr/bin/newaliases.postfix
queue_directory = /var/spool/postfix
readme_directory = /usr/share/doc/postfix-2.3.3/README_FILES
relay_domains = $mydestination
sample_directory = /usr/share/doc/postfix-2.3.3/samples
sendmail_path = /usr/sbin/sendmail.postfix
setgid_group = postdrop
smtpd_client_restrictions = permit_mynetworks, check_client_access hash:/etc/postfix/client_access, permit
smtpd_helo_required = yes
smtpd_helo_restrictions =
smtpd_recipient_restrictions = permit_mynetworks, permit_sasl_authenticated, reject_unauth_destination, reject_invalid_hostname, reject_non_fqdn_hostname, reject_unauth_pipelining, reject_non_fqdn_sender, reject_unknown_sender_domain, reject_unknown_recipient_domain, reject_rbl_client bl.spamcop.net, permit
smtpd_sender_restrictions =
strict_rfc821_envelopes = yes
unknown_local_recipient_reject_code = 550

 

[CXliving]

Hi again guys.
Just a foot note - I swapped the current Bay router for my old but trusty Intel router, and it made no difference at all. However, I have noticed that mail has started flowing through again albeit small ones. larger emails (well, larger being 30k+) are still being dropped.
However, the error now seldom ever has a (0) in it - now it's always 30k+ in brackets. Some larger stuff is coming through, but anything smaller than 30k comes through in under a second.
Thanks again,
Rex