"timeout after DATA from ..."

Question

"timeout after DATA from ..."

I am noticing this message in mail.log quite a bit and wondering if it is indicative of a problem.

Sep 28 13:38:05 Xserve-2 postfix/smtpd[7766]: timeout after DATA from cnnimail14.cnn.com[64.236.25.99]
Sep 28 13:38:05 Xserve-2 postfix/smtpd[7766]: disconnect from cnnimail14.cnn.com[64.236.25.99]

(The message occurs for other domains as well.)

jdis$ postconf -n
alias_maps = hash:/etc/aliases,hash:/var/mailman/data/aliases
command_directory = /usr/sbin
config_directory = /etc/postfix
content_filter = smtp-amavis:[127.0.0.1]:10024
daemon_directory = /usr/libexec/postfix
debug peerlevel = 2
html_directory = no
inet_interfaces = all
mail_owner = postfix
mailbox sizelimit = 0
mailbox_transport = cyrus
mailq_path = /usr/bin/mailq
manpage_directory = /usr/share/man
mydestination = $myhostname,localhost.$mydomain,jdis.com
mydomain = jdis.com
mydomain_fallback = localhost
myhostname = jdis.com
mynetworks = 127.0.0.1/32,206.74.67.64/32,jdis.com,citcom.net
mynetworks_style = host
newaliases_path = /usr/bin/newaliases
owner requestspecial = no
queue_directory = /private/var/spool/postfix
readme_directory = /usr/share/doc/postfix
recipient_delimiter = +
sample_directory = /usr/share/doc/postfix/examples
sendmail_path = /usr/sbin/sendmail
setgid_group = postdrop
smtpd tls_keyfile =
unknown local_recipient_rejectcode = 550

Posted on Sep 28, 2005 11:49 AM

Reply

Answer 1

pterobyte

Level 6

11,104 points

Oct 3, 2005 3:41 AM in response to Clark Wilkins1

As long as your mail is being sent and delivered properly and the disconnetcion messages are just a few, you most likely needn't worry.

It can be inactivity. ISP problems between sender and recipient. Incorrect HELO, etc....

Reply

Answer 2

Oct 4, 2005 1:42 PM in response to pterobyte

It is happening on many accounts. We have a client in Florida who cannot get through to us at all. Here is the excerpt from their particular transaction,

Oct 4 06:46:19 Xserve-2 postfix/smtpd[27738]: connect from 65.106.9.219.ptr.us.xo.net[65.106.9.219]
Oct 4 06:46:19 Xserve-2 postfix/smtpd[27738]: D0EBB199FC1: client=65.106.9.219.ptr.us.xo.net[65.106.9.219]
...
Oct 4 06:51:19 Xserve-2 postfix/smtpd[27738]: timeout after DATA from 65.106.9.219.ptr.us.xo.net[65.106.9.219]
Oct 4 06:51:19 Xserve-2 postfix/smtpd[27738]: disconnect from 65.106.9.219.ptr.us.xo.net[65.106.9.219]

Thanks for the assistance Alex.

Reply

Answer 3

pterobyte

Level 6

11,104 points

Oct 4, 2005 2:14 PM in response to Clark Wilkins1

Hi Clark,

Looks either like a networking issue or a problem with that specific client in Florida.

I initiated a session to your server at 3:06 pm (discussions time) and left it hanging, so you should see a disconnect sooner or later.

Let me know by e-mail (send me the whole log bit from my connection to the disconnect).

Alex

Reply

Answer 4

Oct 8, 2005 6:06 AM in response to pterobyte

No word back from you Alex (or anyone else). This problem continues unabated. We have several clients who tell us that we are unreachable from their end.

Reply

Answer 5

Oct 8, 2005 7:11 AM in response to Clark Wilkins1

OK. I rolled up my sleeves and got further into this. (I have avoided that till now because mail/network stuff is NOT my area of expertise.)

This page references the same problem:

http://archives.neohapsis.com/archives/postfix/2004-01/1899.html

Postfix.org also talks about it here:

http://www.postfix.org/faq.html#timeouts

Mac OSX Hints talks about changing the MTU parameter here:

http://www.macosxhints.com/article.php?story=2005082115321647

But my new question is whether the MTU setting can really be influenced in this manner. The Xserve is connected over DLS (PPoE) via an Asante router. It seems to me that the router might be the determining factor?

Reply

Answer 6

Oct 8, 2005 7:30 AM in response to Clark Wilkins1

I made the changes suggested in macosxhints.com on the off-chance. No change as can be seen in this log extract:

Oct 8 09:25:55 Xserve-2 postfix/smtpd[366]: timeout after DATA from sdm3.quickbooks.net[208.240.241.110]
Oct 8 09:25:55 Xserve-2 postfix/smtpd[366]: disconnect from sdm3.quickbooks.net[208.240.241.110]
Oct 8 09:26:42 Xserve-2 postfix/smtpd[451]: connect from invest.uptownclinic.net[66.201.85.22]
Oct 8 09:26:42 Xserve-2 postfix/smtpd[451]: 70F971BDC1C: client=invest.uptownclinic.net[66.201.85.22]
Oct 8 09:26:42 Xserve-2 postfix/cleanup[357]: 70F971BDC1C: message-id=<1011.168933970065@invest.uptownclinic.net>
Oct 8 09:26:42 Xserve-2 postfix/qmgr[279]: 70F971BDC1C: from=<Tech@uptownclinic.net>, size=2602, nrcpt=1 (queue active)
Oct 8 09:26:42 Xserve-2 postfix/smtpd[451]: disconnect from invest.uptownclinic.net[66.201.85.22]
Oct 8 09:26:44 Xserve-2 postfix/smtpd[496]: connect from localhost[127.0.0.1]
Oct 8 09:26:44 Xserve-2 postfix/smtpd[496]: C21CC1BDC38: client=localhost[127.0.0.1]
Oct 8 09:26:44 Xserve-2 postfix/cleanup[372]: C21CC1BDC38: message-id=<1011.168933970065@invest.uptownclinic.net>
Oct 8 09:26:44 Xserve-2 postfix/smtpd[496]: disconnect from localhost[127.0.0.1]
Oct 8 09:26:44 Xserve-2 postfix/qmgr[279]: C21CC1BDC38: from=<Tech@uptownclinic.net>, size=3163, nrcpt=1 (queue active)
Oct 8 09:26:44 Xserve-2 postfix/smtp[493]: 70F971BDC1C: to=<clark@jdis.com>, relay=127.0.0.1[127.0.0.1], delay=2, status=sent (250 2.6.0 Ok, id=00138-04, from MTA: 250 Ok: queued as C21CC1BDC38)
Oct 8 09:26:44 Xserve-2 postfix/qmgr[279]: 70F971BDC1C: removed
Oct 8 09:26:45 Xserve-2 postfix/pipe[498]: C21CC1BDC38: to=<clark@jdis.com>, relay=cyrus, delay=1, status=sent (jdis.com)
Oct 8 09:26:45 Xserve-2 postfix/qmgr[279]: C21CC1BDC38: removed
Oct 8 09:26:54 Xserve-2 postfix/smtpd[359]: timeout after DATA from mxpool15.ebay.com[66.135.197.21]
Oct 8 09:26:54 Xserve-2 postfix/smtpd[359]: disconnect from mxpool15.ebay.com[66.135.197.21]
Oct 8 09:27:40 Xserve-2 postfix/smtpd[438]: timeout after DATA from 65.106.9.219.ptr.us.xo.net[65.106.9.219]

I also looked at the docs for the router (Asante FR3004) and found nothing to adjust the MTU in case it's on that side.

😟

Reply

Answer 7

pterobyte

Level 6

11,104 points

Oct 8, 2005 9:20 AM in response to Clark Wilkins1

I sent you an e-mail on Oct. 5th because you had sent me the wrong log timeframe. You never replied. 😉

Alex

Reply

Answer 8

pterobyte

Level 6

11,104 points

Oct 9, 2005 3:04 AM in response to pterobyte

I just responded to your lates mail. check if you got it please.

Alex

Reply

Answer 9

Oct 9, 2005 6:35 AM in response to Clark Wilkins1

After much googling, I think I found the answer.

Using this command:

%sudo networksetup -setMTU 'Built-in Ethernet 1' 1450

and rebooting seemed to fix the problem. I have a few E-mails that are several days old starting to dribble in now (from ebay.com which was one of my problem accounts).

This problem is discussed on postfix.org at:

http://www.postfix.org/faq.html#timeouts

Reply

Answer 10

Oct 11, 2005 10:26 AM in response to Clark Wilkins1

Success all around. Our mail is working fine again.

btw: Postfix.org suggests this is a sender problem.

We found it unlikely that we'd be able to convince domains such as ebay.com and cnn.com (just to name two) to reconfigure their side so our side could receive their E-mails. 😉

Reply