Dear Postfix Experts:

I am having difficulties delivering newsletters to comcast.net email
addresses and ask for some advice.  I have set aside a special transport
for the IP (using sender_transport) as it is new without an established
reputation.  When a queue run is begun, comcast.net will allow 50 messages
to be sent before it begins deferring (see message below).  The trouble is
that Postfix only genuinely attempts to deliver messages to comcast.net
every 50 minutes; the queue manager says that it attempts deleveries every
5 minutes, but I think it is just repeating the deferral messages that it
already received.  It is almost as if the minimal_backoff_time that was
defined in the custom transport is not overriding the default setting in
main.cf.

I would like to have it make a 'genuine' attempt more frequently.  With a
deferred queue of 6,000 messages, this will take forever.  Any suggestions
would be greatly appreciated.  Thanks!

Best, Steve


The default configuration is overridden by a custom transport in master.cf
as below.  The postfix version is 2.9.4

------------------------------------
$>postconf -c /etc/postfix-br -n

authorized_submit_users =
command_directory = /usr/sbin
config_directory = /etc/postfix-br
daemon_directory = /usr/libexec/postfix
data_directory = /var/lib/postfix-br
debug_peer_level = 2
debugger_command = PATH=/bin:/usr/bin:/usr/local/bin:/usr/X11R6/bin ddd
$daemon_directory/$process_name $process_id & sleep 5
default_destination_concurrency_limit = 20
default_process_limit = 230
fast_flush_domains = mydomain.net
html_directory = no
inet_protocols = ipv4
mail_owner = postfix
mailq_path = /usr/bin/mailq.postfix
manpage_directory = /usr/share/man
master_service_disable =
maximal_queue_lifetime = 5d
minimal_backoff_time = 45m
multi_instance_enable = yes
multi_instance_name = postfix-br
newaliases_path = /usr/bin/newaliases.postfix
queue_directory = /var/spool/postfix-br
readme_directory = /usr/share/doc/postfix-2.6.6/README_FILES
sample_directory = /usr/share/doc/postfix-2.6.6/samples
sender_dependent_default_transport_maps =
hash:/etc/postfix-br/sender_transport.txt
sendmail_path = /usr/sbin/sendmail.postfix
setgid_group = postdrop
smtp_bind_address = 185.64.153.201
smtp_helo_timeout = 100s
syslog_name = br-shared
unknown_local_recipient_reject_code = 550
virtual_alias_maps =
hash:/var/custom/virtual.txt,regexp:/var/listpilot/virtual_custom.regexp


------------------------------------
$>cat /etc/postfix-br/master.cf

...
acct_client1        unix  -   -      n       -       -       smtp
  -o syslog_name=br-client1
  -o smtp_bind_address=185.94.24.84
  -o smtp_helo_name=news.client1.org
  -o smtp_destination_concurrency_limit=2
  -o smtp_destination_recipient_limit=5
  -o smtp_destination_rate_delay=1s
  -o initial_destination_concurrency=2
  -o minimal_backoff_time=300s
  -o acct_client1_destination_concurrency_failed_cohort_limit=200
...


-------------------------------------
Sample log Message when throttling kicks in:

May  2 16:37:25 mail4 br-client1/smtp[4599]: 2277ACEC472: to=<
pers...@comcast.net>, relay=mx1.comcast.net[96.114.157.80]:25,
delay=400519, delays=400503/11/4.7/0, dsn=4.0.0, status=deferred (host
mx1.comcast.net[96.114.157.80] refused to talk to me: 421
resimta-po-10v.sys.comcast.net comcast 185.94.24.84 Throttled - try again
later. pYdN1s02k1ors0201 Please see
http://postmaster.comcast.net/smtp-error-codes.php#RL000002)

Reply via email to