[Raw Msg Headers][Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Mailq output error



On Wed, Feb 28, 2001 at 08:03:32PM -0300, zmailer wrote:
> Hi,
> we are running ZMailer scheduler (2.99.55 #1:
> on 3 similar solaris 2.6
> 
> On one of them  after a couple of minutes running,
> mailq starts giving negative numbers for the number of messages in transport
> queue. The other 2 schedulers keep running without problem.
> 
> The only difference is that this zmailer only does local delivery an all
> outgoing to a routedb defined smarthost.
> 
> This negative number decrements almost continuously.. see the report at the
> end.
> 
> The same thing happened with 2.99.54p1.

  This is very old problem, but where, that I haven't been able to
  locate, as I see it so damn rarely myself.

  And it is a beauty-wart, not real system threatening thing, that
  is why I havent' burned very much effort there (of lately).

  Do you see any odd things mentioned at your  $LOGDIR/scheduler ?
  Of the machine which has the decrementing counters ?
  I would like to know which part of the internal book-keeping is
  doing double-free, or some such.  (To have even some *hint* at it!)

> Any idea?
> 
> Thanks.
> 
> By the way, Matti, we have almost no problem running 54 router with 55
> scheduler, ta's and smtpserver. The only thing we had to do is add
> a link from /etc/zmailer.conf
> to $MAILSHARE. All .55 programs core dumped without it. Thanks!

  Damn, I forgot to mention that at the README.UPGRADING.
  I have aspired to move all ZMailer system local files to  $MAILSHARE/
  (if shared by all nodes), and to $MAILVAR/ (if node local).
  This time around the  zmailer.conf  got moved to $MAILSHARE/, and
  mail.conf   to  $MAILVAR/.

> MAILQ -sQQQ output... (same with any Q combination)
> 
> 2 entries in router queue: processing
> 0 messages in transport queue: idle
> Kids: 13  Idle: 13  Msgs:   0  Thrds:   0  Rcpnts:    0  Uptime: 24 sec
> Msgs in 13 out 13 stored 0 Rcpnts in 13 out 13 stored 0
> ....
> 17 entries in router queue: processing
> 3 messages in transport queue: working
> Kids: 17  Idle: 16  Msgs:   3  Thrds:   1  Rcpnts:    3  Uptime: 668 sec
> Msgs in 487 out 484 stored 3 Rcpnts in 552 out 549 stored 3
> ...
> 2 entries in router queue: processing
> 23 messages in transport queue: working
> Kids: 17  Idle: 17  Msgs:  -1  Thrds:   0  Rcpnts:    0  Uptime: 689 sec
> Msgs in 516 out 515 stored 4294967295 Rcpnts in 580 out 580 stored 0
> ...
> 6 entries in router queue: processing
> 21 messages in transport queue: working
> Kids: 17  Idle: 16  Msgs:   1  Thrds:   1  Rcpnts:    2  Uptime: 708 sec
> Msgs in 533 out 530 stored 1 Rcpnts in 603 out 601 stored 2
> ...
> 5 entries in router queue: processing
> 17 messages in transport queue: working
> Kids: 17  Idle: 17  Msgs:  -1  Thrds:   0  Rcpnts:    0  Uptime: 728 sec
> Msgs in 557 out 556 stored 4294967295 Rcpnts in 627 out 627 stored 0
> ...
> 2 entries in router queue: processing
> 6 messages in transport queue: working
> Kids: 17  Idle: 17  Msgs:  -2  Thrds:   0  Rcpnts:    0  Uptime: 1034 sec
> Msgs in 746 out 744 stored 4294967294 Rcpnts in 849 out 849 stored 0
> ...
> 1 entry in router queue: processing
> 2 messages in transport queue: working
> Kids: 17  Idle: 16  Msgs:   0  Thrds:   1  Rcpnts:    2  Uptime: 1054 sec
> Msgs in 760 out 756 stored 0 Rcpnts in 887 out 873 stored 14
> ...
> 
> This samples where taken during off-hours, after a day this number go as low
> as a couple of hundreds...
> 
> Nicolás Baumgarten
> Pert Consultores 
> nico@pert.com.ar

-- 
/Matti Aarnio	<mea@nic.funet.fi>