>My boss is blaming IMGate for his email list noit getting through
>
>(There may be more than 4-5 subscribers in our organziation so I've
>throttled the unauthorized pipelining settings and it didnt change
>anything.)
>
>The logs follow:
>grep-ing dundee.net (their mail server.)
>Jun 24 07:33:06 mailhub postfix/smtpd[8054]: connect from
>caseyjones.dundee.net[216.234.106.37]
>Jun 24 07:33:06 mailhub postfix/smtpd[8054]: 87236982C5:
>client=caseyjones.dundee.net[216.234.106.37]
>Jun 24 07:33:37 mailhub postfix/smtpd[8054]: timeout after RSET from
>caseyjones.dundee.net[216.234.106.37]

why is that server sending a RSET and then doing nothing?

>Jun 24 07:33:37 mailhub postfix/smtpd[8054]: disconnect from
>caseyjones.dundee.net[216.234.106.37]
>Jun 24 09:32:18 mailhub postfix/smtpd[8393]: connect from
>caseyjones.dundee.net[216.234.106.37]
>Jun 24 09:32:19 mailhub postfix/smtpd[8393]: EE057982C5:
>client=caseyjones.dundee.net[216.234.106.37]
>Jun 24 09:32:53 mailhub postfix/smtpd[8393]: timeout after RSET from
>caseyjones.dundee.net[216.234.106.37]
>Jun 24 09:32:53 mailhub postfix/smtpd[8393]: disconnect from
>caseyjones.dundee.net[216.234.106.37]
>
>I dont know who needs to tweak what,
>Does anyone offer me any suggestions what more to check????

postconf | grep timeout

tx1# postconf | grep timeout
daemon_timeout = 18000s
ipc_timeout = 3600s
lmtp_connect_timeout = 0s
lmtp_data_done_timeout = 600s
lmtp_data_init_timeout = 120s
lmtp_data_xfer_timeout = 180s
lmtp_lhlo_timeout = 300s
lmtp_mail_timeout = 300s
lmtp_quit_timeout = 300s
lmtp_rcpt_timeout = 300s
lmtp_rset_timeout = 120s
qmqpd_timeout = 300s
smtp_connect_timeout = 60s
smtp_data_done_timeout = 60s
smtp_data_init_timeout = 60s
smtp_data_xfer_timeout = 60s
smtp_helo_timeout = 60s
smtp_mail_timeout = 60s
smtp_quit_timeout = 60s
smtp_rcpt_timeout = 60s
smtp_rset_timeout = 120s
smtpd_timeout = 20s  <<<<<<<<<<<<
trigger_timeout = 10s

The other machine seems to be screwed up (sending RSET junk for no reason) 
and then very slow to send the next command so postfix times out and hangs 
up.  You'll have to increase postfix timeout SMTPD timeout so postfix waits 
long enough for the screwed machine to do something.

Len


Reply via email to