Yahoo Groups archive

Milter-greylist

Index last updated: 2026-04-28 23:32 UTC

Thread

milter-greylist 2.02 crashes

milter-greylist 2.02 crashes

2006-08-19 by Jim Hermann - UUN Hostmaster

What do these log entired indicate about my problems with milter-greylist.
I am running Sendmail 8.13 or Fedora 2.

Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
(greylist): read returned -1: Connection reset by
ppp-124.120.137.59.revip2.asianet.co.th
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2Kf015057: Milter
(greylist): read returned -1: Connection reset by
ppp-124.120.137.59.revip2.asianet.co.th
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2Kf015057: Milter
(greylist): to error state
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2Kf015057: Milter
(greylist): init failed to open
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2Kf015057: Milter
(greylist): to error state
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
(greylist): to error state
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
(greylist): init failed to open
Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
(greylist): to error state


Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6uo015057: Milter
(greylist): read returned -1: Connection reset by localhost.localdomain
Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6uo015057: Milter
(greylist): to error state
Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6uo015057: Milter
(greylist): init failed to open
Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6uo015057: Milter
(greylist): to error state
Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6Pi015057: Milter
(greylist): write(Q) returned -1, expected 5: Broken pipe
Aug 18 08:20:06 host sm-acceptingconnections[15057]: k7IDK6Pi015057: Milter
(greylist): to error state

-----
Jim Hermann <hostmaster@...>
UUism Networks <http://www.UUism.net>
Ministering to the Needs of Online UUs
Web Hosting, Email Services, Mailing Lists
-----

Re: [milter-greylist] milter-greylist 2.02 crashes

2006-08-19 by manu@netbsd.org

Jim Hermann - UUN Hostmaster <hostmaster@...> wrote:

> What do these log entired indicate about my problems with milter-greylist.
> I am running Sendmail 8.13 or Fedora 2.
> 
> Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
> (greylist): read returned -1: Connection reset by
> ppp-124.120.137.59.revip2.asianet.co.th

A usual issue is that it ran out of memory. You may have to increase
allocatable memory with ulimit before starting up milter-greylist. 

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu@...

Re: [milter-greylist] milter-greylist 2.02 crashes

2006-08-19 by Fabien Tassin

According to manu@...:
> 
> > What do these log entired indicate about my problems with milter-greylist.
> > I am running Sendmail 8.13 or Fedora 2.
> > 
> > Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057: Milter
> > (greylist): read returned -1: Connection reset by
> > ppp-124.120.137.59.revip2.asianet.co.th
> 
> A usual issue is that it ran out of memory. You may have to increase
> allocatable memory with ulimit before starting up milter-greylist. 

hmm, all *alloc()/*dup() calls seem to be assigned with a syslog in case
of failure. Should be visible in the error logs then.

What about running through valgrind or something similar. Whatever the problem
is, it will be loggued for sure, maybe even faster than in real life as
those tools usually allocate more memory than requested in order to check for
over&underflows.

/Fabien

RE: [milter-greylist] milter-greylist 2.02 crashes

2006-08-19 by Jim Hermann - UUN Hostmaster

Which memory do I need to increase?
 
[root@host root]# ulimit -a
core file size        (blocks, -c) 0
data seg size         (kbytes, -d) unlimited
file size             (blocks, -f) unlimited
max locked memory     (kbytes, -l) 32
max memory size       (kbytes, -m) unlimited
open files                    (-n) 1024
pipe size          (512 bytes, -p) 8
stack size            (kbytes, -s) 10240
cpu time             (seconds, -t) unlimited
max user processes            (-u) 16355
virtual memory        (kbytes, -v) unlimited



  _____  
Show quoted textHide quoted text
From: milter-greylist@yahoogroups.com
[mailto:milter-greylist@yahoogroups.com] On Behalf Of manu@...
Sent: Saturday, August 19, 2006 12:05 PM
To: milter-greylist@yahoogroups.com
Subject: Re: [milter-greylist] milter-greylist 2.02 crashes



Jim Hermann - UUN Hostmaster <hostmaster@uuism.
<mailto:hostmaster%40uuism.net> net> wrote:

> What do these log entired indicate about my problems with milter-greylist.
> I am running Sendmail 8.13 or Fedora 2.
> 
> Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057:
Milter
> (greylist): read returned -1: Connection reset by
> ppp-124.120.137.59.revip2.asianet.co.th

A usual issue is that it ran out of memory. You may have to increase
allocatable memory with ulimit before starting up milter-greylist. 

-- 
Emmanuel Dreyfus
http://hcpnet. <http://hcpnet.free.fr/pubz> free.fr/pubz
manu@netbsd. <mailto:manu%40netbsd.org> org


 


--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.405 / Virus Database: 268.11.3/423 - Release Date: 08/18/06

RE: [milter-greylist] milter-greylist 2.02 crashes

2006-08-22 by Jim Hermann - UUN Hostmaster

At the end of this message are all the entries from 2 seconds of my maillog
from a crash today, starting with the last greylist action and ending with
the error message.
 
Prior to the crash, the connecting site ( tried to connect 18 times over a
period of 60 seconds.  The last 9 attempts were made three at a time during
1 second intervals.  The last triple occurred 2 seconds before the failure.
The connected site was h66-201-246-167.gtcust.grouptelecom.net
[66.201.246.167].  
 
Can multiple dropped connections cause a problem with milter-greylist?
 
The last triple connection attempt look like this:
 
Aug 21 15:22:31 host sm-acceptingconnections[21836]: k7LKMOSD021836: lost
input channel f
rom h66-201-246-167.gtcust.grouptelecom.net [66.201.246.167] to MTA after
data
Aug 21 15:22:31 host sm-acceptingconnections[21836]: k7LKMOvX021836: lost
input channel f
rom h66-201-246-167.gtcust.grouptelecom.net [66.201.246.167] to MTA after
data
Aug 21 15:22:31 host sm-acceptingconnections[21836]: k7LKMHbB021836: lost
input channel f
rom h66-201-246-167.gtcust.grouptelecom.net [66.201.246.167] to MTA after
data

Here are the 2 seconds of maillog:
 
Aug 21 15:22:32 host milter-greylist: k7LKMOLk021836: addr 66.201.246.167
from <Marshbuzzing@roteiroturismo
.com.br> to <webmaster@...> delayed for 00:06:00
Aug 21 15:22:32 host sm-acceptingconnections[21836]: k7LKMOLk021836: Milter:
to=<webmaster@...>, rej
ect=451 4.7.1 Greylisting in action, please come back in 00:06:00
Aug 21 15:22:32 host milter-greylist: k7LKMVTt021836: skipping greylist
because this is the default action,
 (from=<>, rcpt=<ucm@...>, addr=153.91.3.98)
Aug 21 15:22:32 host ipop3d[21826]: Logout
user=dre@... host=63-231-157-157.mpls.qwes
t.net [63.231.157.157] nmsgs=0 ndele=0
Aug 21 15:22:32 host sm-acceptingconnections[21836]: k7LKMVTt021836:
from=<>, size=939, class=0, nrcpts=1,
msgid=<s4e9cf30.005@...>, proto=ESMTP, daemon=MTA,
relay=cmsu1gate.cmsu.edu [153.91.3.98]
Aug 21 15:22:32 host sm-acceptingconnections[21836]: k7LKMVTt021836:
to=<ucm@...>, delay=00:00:00,
 mailer=virthostmail, pri=30939, stat=queued
Aug 21 15:22:33 host ipop3d[21812]: Login user=uuce@...
host=adsl-75-2-75-139.dsl.chcgil.sbcglobal.net
 [75.2.75.139] nmsgs=0/0
Aug 21 15:22:33 host ipop3d[21812]: Logout user=uuce@...
host=adsl-75-2-75-139.dsl.chcgil.sbcglobal.ne
t [75.2.75.139] nmsgs=0 ndele=0
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMXud021836: Milter
(greylist): read returned -1: C
onnection reset by h66-201-246-167.gtcust.grouptelecom.net
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMXud021836: Milter
(greylist): to error state
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMXud021836: Milter
(greylist): init failed to open
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMXud021836: Milter
(greylist): to error state
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMX6x021836: Milter
(greylist): error connecting to
 filter: Connection refused by /var/milter-greylist/milter-greylist.sock
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMX6x021836: Milter
(greylist): to error state
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMX6x021836:
akk142.internetdsl.tpnet.pl [83.17.14.
142] did not issue MAIL/EXPN/VRFY/ETRN during connection to MTA
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMOLk021836: Milter
(greylist): write(Q) returned -
1, expected 5: Broken pipe
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMOLk021836: Milter
(greylist): to error state
Aug 21 15:22:33 host sm-acceptingconnections[21836]: k7LKMOLk021836: lost
input channel from h66-201-246-16
7.gtcust.grouptelecom.net [66.201.246.167] to MTA after data

 


  _____  
Show quoted textHide quoted text
From: milter-greylist@yahoogroups.com
[mailto:milter-greylist@yahoogroups.com] On Behalf Of Fabien Tassin
Sent: Saturday, August 19, 2006 12:23 PM
To: milter-greylist@yahoogroups.com
Subject: Re: [milter-greylist] milter-greylist 2.02 crashes



According to manu@netbsd. <mailto:manu%40netbsd.org> org:
> 
> > What do these log entired indicate about my problems with
milter-greylist.
> > I am running Sendmail 8.13 or Fedora 2.
> > 
> > Aug 17 10:14:02 host sm-acceptingconnections[15057]: k7HFE2lN015057:
Milter
> > (greylist): read returned -1: Connection reset by
> > ppp-124.120.137.59.revip2.asianet.co.th
> 
> A usual issue is that it ran out of memory. You may have to increase
> allocatable memory with ulimit before starting up milter-greylist. 

hmm, all *alloc()/*dup() calls seem to be assigned with a syslog in case
of failure. Should be visible in the error logs then.

What about running through valgrind or something similar. Whatever the
problem
is, it will be loggued for sure, maybe even faster than in real life as
those tools usually allocate more memory than requested in order to check
for
over&underflows.

/Fabien


 


--
No virus found in this incoming message.
Checked by AVG Free Edition.
Version: 7.1.405 / Virus Database: 268.11.3/423 - Release Date: 08/18/06

Re: [milter-greylist] milter-greylist 2.02 crashes

2006-08-22 by manu@netbsd.org

Jim Hermann - UUN Hostmaster <hostmaster@...> wrote:

> Can multiple dropped connections cause a problem with milter-greylist?

Not as far as I know.

Run milter-greylist -Dv within gdb and issue a bt command when it
crashes. That will help tracking the problem down.

-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu@...

RE: [milter-greylist] milter-greylist 2.02 crashes

2006-08-25 by Jim Hermann - UUN Hostmaster

I don't have gdb available, so I have been running in foreground with the
-Dv options.  The only crash so far left a message about an Alarm.  

Also, it appears that milter-greylist may be more stable in verbose mode.  I
ran the Daemon for a couple of days in verbose mode without any incidents.

One thing that I have noticed is that my greylist.db is over 3 MB.

Why does milter-greylist keep tuples that were delayed and did not retry
before the greylist_delay transpired?

When I start milter-greylist, it takes several minutes to load the existing
database in verbose mode.

Jim

________________________________
Show quoted textHide quoted text
	From: milter-greylist@yahoogroups.com
[mailto:milter-greylist@yahoogroups.com] On Behalf Of manu@...
	Sent: Monday, August 21, 2006 11:45 PM
	To: milter-greylist@yahoogroups.com
	Subject: Re: [milter-greylist] milter-greylist 2.02 crashes
	
	

	Jim Hermann - UUN Hostmaster <hostmaster@...
<mailto:hostmaster%40uuism.net> > wrote:
	
	> Can multiple dropped connections cause a problem with
milter-greylist?
	
	Not as far as I know.
	
	Run milter-greylist -Dv within gdb and issue a bt command when it
	crashes. That will help tracking the problem down.
	
	-- 
	Emmanuel Dreyfus
	http://hcpnet.free.fr/pubz <http://hcpnet.free.fr/pubz> 
	manu@... <mailto:manu%40netbsd.org>

Re: [milter-greylist] milter-greylist 2.02 crashes

2006-08-25 by manu@netbsd.org

Jim Hermann - UUN Hostmaster <hostmaster@...> wrote:

> I don't have gdb available, so I have been running in foreground with the
> -Dv options.  The only crash so far left a message about an Alarm.  
> 
> Also, it appears that milter-greylist may be more stable in verbose mode.  I
> ran the Daemon for a couple of days in verbose mode without any incidents.

That suggests a timing-dependent issue, since the olny difference is
that it gets slower due to debug printf.

> One thing that I have noticed is that my greylist.db is over 3 MB.
> 
> Why does milter-greylist keep tuples that were delayed and did not retry
> before the greylist_delay transpired?

When we cleanup any tuple, we have to walk the database, and we use that
opportunity to cleanup any expired entry we find.

> When I start milter-greylist, it takes several minutes to load the existing
> database in verbose mode.

You can remove the printf related to database laoding, that will speed
it up.


-- 
Emmanuel Dreyfus
http://hcpnet.free.fr/pubz
manu@...

RE: [milter-greylist] milter-greylist 2.02 crashes

2006-08-25 by Jim Hermann - UUN Hostmaster

I decreased the timeout from 5 days to 26 hours for keeping tuples.  My
database decreased to 400 KB.

What's the point in keeping tuples for so long?

If my greylist_delay is 6 minutes, I would think we could discard tuples
after 10 minutes, assuming that timeout does not affect the time for
auto-white-listing.

Jim

________________________________
Show quoted textHide quoted text
	From: milter-greylist@yahoogroups.com
[mailto:milter-greylist@yahoogroups.com] On Behalf Of manu@...
	Sent: Friday, August 25, 2006 12:20 AM
	To: milter-greylist@yahoogroups.com
	Subject: Re: [milter-greylist] milter-greylist 2.02 crashes
	
	
	
	Jim Hermann - UUN Hostmaster <hostmaster@...
<mailto:hostmaster%40uuism.net> > wrote:
	
	> I don't have gdb available, so I have been running in foreground
with the
	> -Dv options. The only crash so far left a message about an Alarm. 
	> 
	> Also, it appears that milter-greylist may be more stable in
verbose mode. I
	> ran the Daemon for a couple of days in verbose mode without any
incidents.
	
	That suggests a timing-dependent issue, since the olny difference is
	that it gets slower due to debug printf.
	
	> One thing that I have noticed is that my greylist.db is over 3 MB.
	> 
	> Why does milter-greylist keep tuples that were delayed and did not
retry
	> before the greylist_delay transpired?
	
	When we cleanup any tuple, we have to walk the database, and we use
that
	opportunity to cleanup any expired entry we find.

Re: [milter-greylist] milter-greylist 2.02 crashes

2006-08-25 by Emmanuel Dreyfus

On Fri, Aug 25, 2006 at 10:54:07AM -0500, Jim Hermann - UUN Hostmaster wrote:
> I decreased the timeout from 5 days to 26 hours for keeping tuples.  My
> database decreased to 400 KB.
> 
> What's the point in keeping tuples for so long?

A legitimate site may go offline for more than a day...

But the setting is there so that you can tweak it to your taste.

-- 
Emmanuel Dreyfus
manu@...

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.