I found strange information in the log:
$ cat /var/log/mail/all|grep 9564
Jun 22 13:01:02 sendmail[9564]: NOQUEUE: connect from [192.168.32.113]
Jun 22 13:01:02 sendmail[9564]: AUTH: available mech=SRP DIGEST-MD5 OTP CRAM-MD5 NTLM PLAIN LOGIN ANONYMOUS, allowed
mech=(null)
Jun 22 13:01:07 sendmail[9564]: p5M912Tv009564: --- 220 gw ESMTP Sendmail 8.14.4/8.14.4; Wed, 22 Jun 2011 13:01:02 +0400
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: <-- EHLO Station1
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-gw Hello [192.168.32.113], pleased to meet you
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-ENHANCEDSTATUSCODES
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-PIPELINING
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-EXPN
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-VERB
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-8BITMIME
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-SIZE 500000000
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-DSN
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-ETRN
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250-DELIVERBY
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: --- 250 HELP
Jun 22 13:01:08 sendmail[9564]: p5M912Tv009564: <-- MAIL FROM:<q...@my.dom> SIZE=584
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: Milter (milter-mailfromd): timeout before data read, where=mail
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: Milter (milter-mailfromd): to error state
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: --- 250 2.1.0 <q...@my.dom>... Sender ok
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: <-- RSET
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: --- 250 2.0.0 Reset state
Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: from=<q...@my.dom>, size=584, class=0, nrcpts=0, proto=ESMTP, daemon=MTA,
relay=[192.168.32.113]
Jun 22 13:02:08 sendmail[9564]: p5M912Tw009564: --- 421 4.4.1 gw Lost input channel from [192.168.32.113]
Jun 22 13:02:08 sendmail[9564]: p5M912Tw009564: lost input channel from [192.168.32.113] to MTA after rset
Last two lines have another queue id. I can not understand it.
Be may a bug here ? This is all log lines with PID 9564 of sendmail.
--
Regards, Sergey.
> Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: --- 250 2.1.0 <q...@my.dom>... Sender ok
> Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: <-- RSET
> Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: --- 250 2.0.0 Reset state
> Jun 22 13:02:08 sendmail[9564]: p5M912Tv009564: from=<q...@my.dom>, size=584, class=0, nrcpts=0, proto=ESMTP, daemon=MTA,
> relay=[192.168.32.113]
> Jun 22 13:02:08 sendmail[9564]: p5M912Tw009564: --- 421 4.4.1 gw Lost input channel from [192.168.32.113]
> Jun 22 13:02:08 sendmail[9564]: p5M912Tw009564: lost input channel from [192.168.32.113] to MTA after rset
>
> Last two lines have another queue id. I can not understand it.
After a ReSET a new transaction starts, hence, it needs to get an
unique queueid. The from=<q.. is the last line of the previous
transaction.
Regards, ska
> After a ReSET a new transaction starts, hence, it needs
> to get an unique queueid.
Thanks ! I had not thought about it.
--
Regards, Sergey
Why is that necessary if it knows that there's nothing queued under the ID
abandoned and unused?
If I were a sendmail developer, I would handle events sequentially:
. There is a RSET commands, hence, purge internal state and open new
transaction
. connection closed, hence, log "lost input channel"