[Mimedefang] Getting error MXCommand: read: Socket is not connected: Is multiplexor running?

Kotis, Ione Ione.Kotis at sinclair.edu
Mon Jul 24 19:32:29 EDT 2006


Having problem with these errors several times a day on my mail relays:


MXCommand: read: Socket is not connected: Is multiplexor running?
mfconnect: Error communicating with multiplexor
Milter: connect: host=[206.161.64.136], addr=206.161.64.136, temp failing
commands


It seems to correlate with discarding spam but I can't pin it down.  We
discard anything scoring >5 and provide users with a digest of what was
discarded. We use /etc/mail/access to limit incoming mail to existing users.

User mailboxes are stored on backend exchange servers.

I've done a number of things, including running with -l and changing
timeouts for MX_BUSY and the milter timeouts in sendmail.mc. Increasing them
seemed to help some.  I'm not running out of memory, file descriptors,
inodes, or disk space. It's not happening on large emails, virus scans
aren't timing out.

Having this happen a couple times a day isn't a big deal unless it's a
indication of a bigger problem.  If all mail tempfailed for a significant
period of time I would be in trouble.   After the upgrade to SA 3.1.2 I had
an SA permissions problem which I didn't catch which caused everything to
tempfail for about 12 hours, and our director threatened me with an
appliance.  "Maybe we're not spending enough money on this.  Maybe we need
an appliance." (Uh, no, maybe we need our NOC staffed 24x7).   So I wrote a
script that detects errors and I get paged when they happen.

I would appreciate any input, especially a sanity check--do my timeout
settings make sense?

Can anyone suggest anything else to look at?


More information below.


Thanks, Ione





MD 2.57
SA 3.1.3
perl5.8.8
clamav 0.88.3
uvscan 5.10.0
bind 9.3.2
apache 2.2 (website for mailstats used by 2-3 admins)
FreeBSD 5.5 and 5.4 on HP DL380 G4 3.4Ghz 4GB RAM




md-mx-ctrl hload
  Load              Msgs        Msgs/Sec     Avg ms/Scan   Avg Busy Slaves
    1h             246            0.72           579.8            1.38
    4h            7455            0.67           642.0            1.57
   12h           19682            0.49           612.9            1.44
   24h           34103            0.41           601.8            1.38





from maillog -----------------------------------

Jul 24 17:10:09 mm mimedefang-multiplexor[563]: Killing idle slave 1 (pid
24409): Slave has processed 300 requests
Jul 24 17:10:09 mm mimedefang-multiplexor[563]: Slave 1 resource usage:
req=300, scans=93, user=38.955, sys=13.058, nswap=0, majflt=0,
minflt=454689, maxrss=31080, bi=0, bo=525
.
.
.

Jul 24 17:10:17 mm sm-mta[26577]: k6OLA71Y026577:
from=<bonnie.sharp.j40q at statefarm.com>, size=997, class=0, nrcpts=1,
msgid=f7sIMNBHKwJow.rwG2Gsoi1VfbD at statefarm.com, proto=SMTP, daemon=MTA,
relay=c-67-176-205-127.hsd1.il.comcast.net [67.176.205.127]
Jul 24 17:10:18 mm mimedefang.pl[25633]:
MDLOG,k6OLA71Y026577,scc-digest,6.754,67.176.205.127,<bonnie.sharp.j40q at stat
efarm.com>,<bonnie.shane at OURDOMAIN.TLD>,P**** E******* P**** works for
everybody: black%2C white%2C and yellow%2C
Jul 24 17:10:18 mm sm-mta[26577]: k6OLA71Y026577: Milter: data, discard
Jul 24 17:10:18 mm sm-mta[26577]: k6OLA71Y026577: discarded

Jul 24 17:10:18 mm sm-mta[26462]: k6OL987S026462: lost input channel from
71-213-209-91.sxcy.qwest.net [71.213.209.91] to MTA after mail
Jul 24 17:10:18 mm sm-mta[26462]: k6OL987S026462:
from=<reggislund at infobel.com>, size=0, class=0, nrcpts=0, proto=ESMTP,
daemon=MTA, relay=
71-213-209-91.sxcy.qwest.net [71.213.209.91]

Jul 24 17:10:19 mm sm-mta[26437]: k6OL8ahr026423:
to=<mikita.bush at tns-global.com>, delay=00:01:33, xdelay=00:00:03,
mailer=esmtp, pri=27408
6, relay=mail.3dns.tns-global.com. [208.46.47.28], dsn=2.0.0, stat=Sent
(B44c53a260000 Message accepted for delivery)



Jul 24 17:10:19 mm mimedefang[579]: MXCommand: read: Socket is not
connected: Is multiplexor running?
Jul 24 17:10:19 mm mimedefang[579]: mfconnect: Error communicating with
multiplexor
Jul 24 17:10:19 mm sm-mta[26593]: k6OLAJp8026593: Milter: connect:
host=[206.161.64.136], addr=206.161.64.136, temp failing commands



end from maillog --------------------------------




from debug.log ----------------------------------

Jul 24 17:09:24 mm mimedefang[579]: k6OL9DMp026469: Discarding because
filter instructed us to
Jul 24 17:10:09 mm mimedefang-multiplexor[563]: Reap: slave 1 (pid 24409)
exited normally with status 0
Jul 24 17:10:18 mm mimedefang.pl[25633]: filter: k6OLA71Y026577:  discard=1
Jul 24 17:10:18 mm mimedefang[579]: k6OLA71Y026577: Discarding because
filter instructed us to
Jul 24 17:10:39 mm mimedefang.pl[25633]: filter: k6OLAREg026597:  discard=1
Jul 24 17:10:39 mm mimedefang[579]: k6OLAREg026597: Discarding because
filter instructed us to


end from debug.log ------------------------------



from init script

MX_RELAY_CHECK=yes
MX_SENDER_CHECK=yes
MX_LOG=yes
MX_SLAVE_DELAY=3
MX_MIN_SLAVE_DELAY=0
MX_REQUESTS=300
MX_MINIMUM=2
MX_MAXIMUM=50
MX_BUSY=1140


sendmail.mc --------------------------------

FEATURE(greet_pause,`10000')dnl 10 seconds
.
.
.
define(`confTO_IDENT', `0s')
define(`confTO_COMMAND', `5m')


dnl INPUT_MAIL_FILTER(`mimedefang',
`S=unix:/var/spool/MIMEDefang/mimedefang.sock, F=T, T=S:5m;R:5m;E:5m')
dnl INPUT_MAIL_FILTER(`mimedefang',
`S=unix:/var/spool/MIMEDefang/mimedefang.sock, F=T, T=C:6m;S:6m;R:6m;E:12m')
dnl INPUT_MAIL_FILTER(`mimedefang',
`S=unix:/var/spool/MIMEDefang/mimedefang.sock, F=T,
T=C:10m;S:10m;R:10m;E:20m')
INPUT_MAIL_FILTER(`mimedefang',
`S=unix:/var/spool/MIMEDefang/mimedefang.sock, F=T,
T=C:20m;S:20m;R:20m;E:20m')



-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 3386 bytes
Desc: not available
URL: <https://lists.mimedefang.org/pipermail/mimedefang_lists.mimedefang.org/attachments/20060724/a122ec82/attachment.bin>


More information about the MIMEDefang mailing list