logo       

RE: return_job and the difference between imsimta view database | directory: msg#00054

mail.ims.general

Subject: RE: return_job and the difference between imsimta view database | directory

The reported behaviour of the qm command at restart is strange. At start
up, the job controller rebuilds its knowledge of the queues is gleamed from
the files queued on disk: that is the only source of information for it.
However, it does do a lot more than qm sum to rebuild the info. So an
obvious question to ask is how soon after the restart did you issue these
commands? Perhaps the job controller had not yet read all the files...

Alternatively, there might be something "wrong" with the supernumerary
files. If so, something would be logged about it in the job controller log
file.

I'm not sure what you mean about tcp_local thinking anything. Tcp_local
does not think much about the queues. The job controller does. The channel
master programs do what the job controller tells them to do.

Beyond that, the next thing to do is to turn on job controller debugging to
see what is really going on. Trouble is that that produces a lot of
information.

Peter Coates

-----Original Message-----
From: Info-iMS-owner-QMRIvgJGioDQT0dZR+AlfA@xxxxxxxxxxxxxxxx
[mailto:Info-iMS-owner-QMRIvgJGioDQT0dZR+AlfA@xxxxxxxxxxxxxxxx] On Behalf
Of donald richardson
Sent: October 17, 2007 09:32
To: Info-iMS-QMRIvgJGioDQT0dZR+AlfA@xxxxxxxxxxxxxxxx
Subject: [Info-iMS] return_job and the difference between imsimta view
database | directory

I have a large disparity in the number of messages my mta reports it has
to deliver. tcp_local thinks that
the oldest mail on the filesystem has an internal date of Sep 15.
However, the database says everything is current.
Sometimes, I can do a read on these old files, other times the messages
do not exist on the filesystem.

This leads me to believe return_job is not doing its job, but after
turning on RETURN_DEBUG=1 in the option.dat
file I get:
Inquiring job_controller for messages requiring notification...
All messages requiring notification at this time processed.

The defaults for all my channels include:
!
! part II : channel blocks
!
defaults logging notices 1 2 3 4 nowarnpost nosendpost postheadonly
noswitchchannel immnonurgent maxjobs 7

So I should have no mail older than 4 days. Out of my 30+ mtas's only 3
of my heavy duty mta's "appear" to keep get rid of all mail and bounces
after 4 days. The rest need manual intervention.


It is believed these "nonexistent" and "undying" messages affect the
performance of my mta's regardless of what the database believes it has
to work with. Which view is telling the truth and why doesn't
return_job kill the seemingly old files? Any ideas would be greatly
appreciated.

Donald Richardson
Verizon Messaging

./imsimta version
Sun Java(tm) System Messaging Server 6.2-6.01 (built Apr 3 2006)
libimta.so 6.2-6.01 (built 11:20:35, Apr 3 2006)
SunOS vms169121 5.9 Generic_118558-39 sun4u sparc SUNW,Sun-Fire-V890

Below is a summary command for view directory, then view database, after
a complete restart of the system.

qm.maint> sum
Messages
Channel Queued Size (Kb) Oldest
-------------------------------- -------- ----------- -----------------
tcp_yahoo 11 119.63 15 Oct, 18:55:56
tcp_netscape 0 0.00
tcp_msn 0 0.00
tcp_local 73701 2879407.01 15 Sep, 05:27:24
tcp_lmtp_resst9v 0 0.00
tcp_lmtp_resst91v 0 0.00
tcp_lmtp_resst8v 0 0.00
tcp_lmtp_resst5v 0 0.00
tcp_lmtp_resst4v 0 0.00
tcp_lmtp_resst3v 0 0.00
tcp_lmtp_resst2v 0 0.00
tcp_lmtp_resst236v 1 2.66 17 Oct, 11:06:04
tcp_lmtp_resst1v 0 0.00
tcp_lmtp_resst190v 5 32.11 17 Oct, 11:06:04
tcp_lmtp_resst189v 3 335.44 17 Oct, 11:06:04
tcp_lmtp_resst188v 5 40.51 17 Oct, 11:06:04
tcp_lmtp_resst187v 0 0.00
tcp_lmtp_resst186v 2 33.36 17 Oct, 11:06:04
tcp_lmtp_resst185v 3 1269.22 17 Oct, 11:06:04
tcp_lmtp_resst168063v 0 0.00
tcp_lmtp_resst168061v 0 0.00
tcp_lmtp_resst168025v 0 0.00
tcp_lmtp_resst168023v 0 0.00
tcp_lmtp_resst168021v 0 0.00
tcp_lmtp_resst168019v 1 2.94 17 Oct, 11:06:04
tcp_lmtp_resst168015v 0 0.00
tcp_lmtp_resst168013v 0 0.00
tcp_lmtp_resst168011v 3 36.18 17 Oct, 11:06:04
tcp_lmtp_resst168009v 0 0.00
tcp_lmtp_resst166v 3 30.18 17 Oct, 11:06:04
tcp_lmtp_resst165v 0 0.00
tcp_lmtp_resst12v 0 0.00
tcp_lmtp_resst11v 0 0.00
tcp_lmtp_resst10v 0 0.00
tcp_intranet 2 13.44 17 Oct, 10:21:00
tcp_hotmail 0 0.00
tcp_gmail 0 0.00
tcp_earthlink 4 445.10 14 Oct, 05:23:48
tcp_aol 0 0.00
reprocess 182 17904.18 12 Oct, 01:30:20
process 15 97.53 17 Oct, 11:06:00
hold 0 0.00
-------------------------------- -------- ----------- -----------------
Totals 73941 2899769.50
qm.maint> view database
View set to database
qm.maint> sum
Messages
Channel Queued Size (Kb) Oldest
-------------------------------- -------- ----------- -----------------
tcp_yahoo 3 27.00 17 Oct, 11:05:03
tcp_netscape 0 0.00
tcp_msn 0 0.00
tcp_local 13851 340221.00 17 Oct, 09:32:56
tcp_lmtp_resst9v 0 0.00
tcp_lmtp_resst91v 0 0.00
tcp_lmtp_resst8v 0 0.00
tcp_lmtp_resst5v 0 0.00
tcp_lmtp_resst4v 0 0.00
tcp_lmtp_resst3v 0 0.00
tcp_lmtp_resst2v 0 0.00
tcp_lmtp_resst236v 0 0.00
tcp_lmtp_resst1v 0 0.00
tcp_lmtp_resst190v 5 58.00 17 Oct, 11:07:05
tcp_lmtp_resst189v 0 0.00
tcp_lmtp_resst188v 2 271.00 17 Oct, 11:07:04
tcp_lmtp_resst187v 1 6.00 17 Oct, 11:07:04
tcp_lmtp_resst186v 8 116.00 17 Oct, 11:07:04
tcp_lmtp_resst185v 0 0.00
tcp_lmtp_resst168063v 0 0.00
tcp_lmtp_resst168061v 0 0.00
tcp_lmtp_resst168025v 0 0.00
tcp_lmtp_resst168023v 1 7.00 17 Oct, 11:07:05
tcp_lmtp_resst168021v 0 0.00
tcp_lmtp_resst168019v 0 0.00
tcp_lmtp_resst168015v 0 0.00
tcp_lmtp_resst168013v 0 0.00
tcp_lmtp_resst168011v 0 0.00
tcp_lmtp_resst168009v 0 0.00
tcp_lmtp_resst166v 1 990.00 17 Oct, 11:07:05
tcp_lmtp_resst165v 18 508.00 17 Oct, 11:07:04
tcp_lmtp_resst12v 0 0.00
tcp_lmtp_resst11v 0 0.00
tcp_lmtp_resst10v 0 0.00
tcp_intranet 0 0.00
tcp_hotmail 1 39.00 17 Oct, 11:07:04
tcp_gmail 0 0.00
tcp_earthlink 0 0.00
tcp_aol 0 0.00
reprocess 0 0.00
process 6 115.00 17 Oct, 11:06:55
hold 0 0.00
-------------------------------- -------- ----------- -----------------
Totals 13897 342358.00
qm.maint>




<Prev in Thread] Current Thread [Next in Thread>
Google Custom Search

News | FAQ | advertise