Looking at the code involved I noticed that the timeout value is never
reset to anything but timeout_login.

So, for low-traffic lmtp installations, set timeout_login in the lmtp
stanza to some value bigger than 60. Please try something like 300.

John Feuerstein wrote:
> With 2.2.13 on Ubuntu Karmic we get lots of:
> 
>> Error:[lmtp] lmtp.c,lmtp_handle_connection(+205): client timed, connection 
>> closed
> 
> We're using these LMTP-related settings:
> 
> [DBMAIL]
> MAXCONNECTS           = 100
> TIMEOUT               = 300
> login_timeout         = 60
> 
> [LMTP]
> PORT                  = 24
> NCHILDREN             = 8
> MAXCHILDREN           = 32
> MINSPARECHILDREN      = 4
> MAXSPARECHILDREN      = 8
> 
> So the settings from DBMAIL section should be inherited. I'm not sure
> why login_timeout is used and not TIMEOUT? Here is a typical snippet out
> of mail.log:
> 
>> Dec  2 16:03:18 mx1 postfix/smtpd[9489]: XXX: client=CCC[NNN]
>> Dec  2 16:03:18 mx1 postfix/cleanup[12000]: XXX: 
>> message-id=<4b167f4e.00000...@ccc>
>> Dec  2 16:03:18 mx1 postfix/qmgr[9688]: XXX: from=<[email protected]>, 
>> size=33584, nrcpt=1 (queue active)
>> Dec  2 16:03:18 mx1 dbmail/lmtpd[9553]: Message:[serverchild] 
>> serverchild.c,PerformChildTask(+349): incoming connection from [127.1.1.3] 
>> by pid [9553]
>> Dec  2 16:03:18 mx1 postfix/smtpd[9489]: disconnect from CCC[NNN]
>> Dec  2 16:03:18 mx1 dbmail/lmtpd[9553]: Message:[sort] 
>> sort.c,sort_deliver_to_mailbox(+196): message id=3227806, size=32835 is 
>> inserted
>> Dec  2 16:03:18 mx1 postfix/lmtp[12001]: XXX: to=<[email protected]>, 
>> orig_to=<[email protected]>, relay=127.1.1.3[127.1.1.3]:24], delay=0.36, 
>> delays=0.25/0/0/0.11, dsn=2.0.0, status=sent (215 Recipient <[email protected]> 
>> OK)
>> Dec  2 16:04:18 mx1 dbmail/lmtpd[9553]: Error:[lmtp] 
>> lmtp.c,lmtp_handle_connection(+205): client timed, connection closed
> 
> Filtering out just dbmail/lmtpd lines:
> 
>> Dec  2 16:03:18 mx1 dbmail/lmtpd[9553]: Message:[serverchild] 
>> serverchild.c,PerformChildTask(+349): incoming connection from [127.1.1.3] 
>> by pid [9553]
>> Dec  2 16:03:18 mx1 dbmail/lmtpd[9553]: Message:[sort] 
>> sort.c,sort_deliver_to_mailbox(+196): message id=3227806, size=32835 is 
>> inserted
>> Dec  2 16:04:18 mx1 dbmail/lmtpd[9553]: Error:[lmtp] 
>> lmtp.c,lmtp_handle_connection(+205): client timed, connection closed
> 
> We can see the timeout happens always after exactly 60 seconds. Message
> delivery is successful, though. Looks like Postfix doesn't get the
> information it needs to close the connection? Or is this close after 60s
> caused by a default Postfix timeout for lmtp?
> 
> This is our related Postfix config:
> 
>> # ==========================================================================
>> # service type  private unpriv  chroot  wakeup  maxproc command + args
>> #               (yes)   (yes)   (yes)   (never) (100)
>> # ==========================================================================
>>
>> dbmail    unix  -       -       n       -       32      lmtp
>>   -o disable_dns_lookups=yes
> 
> So MAXCHILDREN=32 is never exceeded.
> 
> And in main.cf we use the dbmail transport:
> 
>> virtual_transport = dbmail:[127.1.1.3]:24
> 
> 
> Any ideas? Thanks.
> _______________________________________________
> DBmail mailing list
> [email protected]
> http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail
> 


-- 
  ________________________________________________________________
  Paul Stevens                                      paul at nfg.nl
  NET FACILITIES GROUP                     GPG/PGP: 1024D/11F8CD31
  The Netherlands________________________________http://www.nfg.nl
_______________________________________________
DBmail mailing list
[email protected]
http://mailman.fastxs.nl/cgi-bin/mailman/listinfo/dbmail

Reply via email to