Help with rt-mailgate

RT Users
1

Hello,

I have the following setup:

RT 1.04 installed
sendmail 8.9.3-20
apache 1.3.12-2
RT is in /usr/local/rt
RH Linux 6.2

The web interface is working excellent (great app). However, I’ve setup a
mail alias on the rt box for a “test” queue I created using admin-webrt.cgi:

rt: |“/usr/local/rt/bin/rt-mailgate test correspond”

When I send mail to this alias, I get the following bounced back:

Reporting-MTA: dns; reserve.ubet.com
Received-From-MTA: DNS; bellagio.ubet.com
Arrival-Date: Tue, 26 Sep 2000 14:37:44 -0700

Final-Recipient: RFC822; rt@reserve.ubet.com
X-Actual-Recipient: RFC822; |/usr/local/rt/bin/rt-mailgate test
correspond@reserve.ubet.com
Action: failed
Status: 5.5.0
Last-Attempt-Date: Tue, 26 Sep 2000 14:37:44 -0700

----- The following addresses had permanent fatal errors -----
|“/usr/local/rt/bin/rt-mailgate test correspond”
(expanded from: rt@reserve.ubet.com)

----- Transcript of session follows -----
sh: rt-mailgate not available for sendmail programs
554 |“/usr/local/rt/bin/rt-mailgate test correspond”… Service unavailable

AND the following in the maillog

Sep 26 15:07:31 reserve sendmail[9407]: PAA09407: from=ecross@ubet.com,
size=484, class=0, pri=30484, nrcpts=1,
msgid=9DB2A69CAA0DD21186BA00A0C961328C01EE6B91@bellagio.ubet.com,
proto=ESMTP, relay=bellagio.ubet.com [10.10.100.15]
Sep 26 15:07:31 reserve smrsh: uid 8: attempt to use rt-mailgate test
correspond
Sep 26 15:07:31 reserve sendmail[9408]: PAA09407:
to=|“/usr/local/rt/bin/rt-mailgate test correspond”, delay=00:00:00,
xdelay=00:00:00, mailer=prog, stat=Service unavailable
Sep 26 15:07:31 reserve sendmail[9408]: PAA09407: PAA09408: DSN: Service
unavailable
Sep 26 15:07:31 reserve sendmail[9408]: PAA09408: to=ecross@ubet.com,
delay=00:00:00, xdelay=00:00:00, mailer=esmtp, relay=bellagio.ubet.com.
[10.10.100.15], stat=Sent (OK)

Thanks for any help,

-Eric

2

RTFM (http://www.fsck.com/projects/rt/docs/FAQ.html, section 5)

Hello,

I have the following setup:

RT 1.04 installed
sendmail 8.9.3-20
apache 1.3.12-2
RT is in /usr/local/rt
RH Linux 6.2

The web interface is working excellent (great app). However, I’ve setup a
mail alias on the rt box for a “test” queue I created using admin-webrt.cgi:

rt: |“/usr/local/rt/bin/rt-mailgate test correspond”

When I send mail to this alias, I get the following bounced back:

Reporting-MTA: dns; reserve.ubet.com
Received-From-MTA: DNS; bellagio.ubet.com
Arrival-Date: Tue, 26 Sep 2000 14:37:44 -0700

Final-Recipient: RFC822; rt@reserve.ubet.com
X-Actual-Recipient: RFC822; |/usr/local/rt/bin/rt-mailgate test
correspond@reserve.ubet.com
Action: failed
Status: 5.5.0
Last-Attempt-Date: Tue, 26 Sep 2000 14:37:44 -0700

----- The following addresses had permanent fatal errors -----
|“/usr/local/rt/bin/rt-mailgate test correspond”
(expanded from: rt@reserve.ubet.com)

----- Transcript of session follows -----
sh: rt-mailgate not available for sendmail programs
554 |“/usr/local/rt/bin/rt-mailgate test correspond”… Service unavailable

AND the following in the maillog

Sep 26 15:07:31 reserve sendmail[9407]: PAA09407: from=ecross@ubet.com,
size=484, class=0, pri=30484, nrcpts=1,
msgid=9DB2A69CAA0DD21186BA00A0C961328C01EE6B91@bellagio.ubet.com,
proto=ESMTP, relay=bellagio.ubet.com [10.10.100.15]
Sep 26 15:07:31 reserve smrsh: uid 8: attempt to use rt-mailgate test
correspond
Sep 26 15:07:31 reserve sendmail[9408]: PAA09407:
to=|“/usr/local/rt/bin/rt-mailgate test correspond”, delay=00:00:00,
xdelay=00:00:00, mailer=prog, stat=Service unavailable
Sep 26 15:07:31 reserve sendmail[9408]: PAA09407: PAA09408: DSN: Service
unavailable
Sep 26 15:07:31 reserve sendmail[9408]: PAA09408: to=ecross@ubet.com,
delay=00:00:00, xdelay=00:00:00, mailer=esmtp, relay=bellagio.ubet.com.
[10.10.100.15], stat=Sent (OK)

Thanks for any help,

-Eric

rt-users mailing list
rt-users@lists.fsck.com
http://lists.fsck.com/mailman/listinfo/rt-users

Happy Landings,

Jon Detert
Unix System Administrator, Milwaukee School of Engineering
1025 N. Broadway, Milwaukee, Wisconsin 53202

3

There is one point that the FAQ doesn’t mention. If you compile sendmail
from source, instead of installing it from the Redhat Linux rpm, the default
location for the smrsh program directory is “/usr/adm/sm.bin”. This directory
doesn’t exist under RH 6.2 by default ("/usr/adm" doesn’t exist, in fact). You
can either create the directory specifically for the compile of smrsh, or you
can edit …/sendmail-8.X.X/smrsh/smrsh.c to replace “/usr/adm/sm.bin” with
"/etc/smrsh".

Personally, I recommend upgrading to sendmail 8.11, which means compiling the
source code. At least, I’ve not seen a rpm/deb distribution of that version.
I’ve not looked for one, though, either.

Dan Shauver
HortResearch UNIX Dude

4

To List,

We’ve had a particular situation come up a couple times now where someone
"outside" our approved (LDAP) organization, was added as an ‘other’ Cc to
a reply email from RT. That person then tried to do a reply back (into RT)
and RT kicked it out, as it should have. However, it seems RT is *
continuing* to send the error message over and over (days). Is there a “for
how long” or a “X times” setting in rt-mailgate we can modify to stop this
endless error message?

Thanks.

Kenn
LBNL

5

We’ve had a particular situation come up a couple times now where someone “outside” our
approved (LDAP) organization, was added as an ‘other’ Cc to a reply email from RT. That person
then tried to do a reply back (into RT) and RT kicked it out, as it should have. However, it
seems RT is continuing to send the error message over and over (days). Is there a “for how
long” or a “X times” setting in rt-mailgate we can modify to stop this endless error message?

Kenn

You’ve not really provided a lot to go on. Is the original mail still
in a postfix queue and being retried because rt-mailgate returns a
TEMPFAIL error or something else? Without some actual data it is
hard.

-kevin

6

Kevin,

Thanks for the reply. Let me try to me a little more forthcoming on the
info.
We have set up our RT session to use LDAP, and to NOT auto create.
We had a ticket where the owner did a “reply” and added an outside email
address to the “Cc”.
This ‘outside’ person replied to the ticket and was getting
about every hour on the hours or so.
We bounced the RT session and that stopped it.

From what I understand, our mailserver is ‘polled’ by RT to get any new
email sent and pulls it into RT-mailgate. There it tries to do whatever and
is kicked out due to either LDAP or privileges. All this sounds good so far.
From that point, I’m trying to figure out if it is sending itself a copy of
that rejection and causing a loop. We don’t use “ParseCc’s” so we didn’t set
the Regexp in our settings. I’m going to try that and see if that stops the
loop from happening.
That’s all I can figure at the moment.
My questions was oriented around rt-mailgate in hopes of finding out of it
had some sort of re-generative of retry for a number of
times or something that I could set to make it happen only once.

I’m thinking the REGexp setting will resolve the problem.

Kenn
LBNLOn Fri, Aug 6, 2010 at 12:36 PM, Kevin Falcone falcone@bestpractical.comwrote:

On Wed, Aug 04, 2010 at 04:10:46PM -0700, Kenneth Crocker wrote:

We’ve had a particular situation come up a couple times now where
someone “outside” our
approved (LDAP) organization, was added as an ‘other’ Cc to a reply
email from RT. That person
then tried to do a reply back (into RT) and RT kicked it out, as it
should have. However, it
seems RT is continuing to send the error message over and over (days).
Is there a “for how
long” or a “X times” setting in rt-mailgate we can modify to stop this
endless error message?

Kenn

You’ve not really provided a lot to go on. Is the original mail still
in a postfix queue and being retried because rt-mailgate returns a
TEMPFAIL error or something else? Without some actual data it is
hard.

-kevin

Discover RT’s hidden secrets with RT Essentials from O’Reilly Media.
Buy a copy at http://rtbook.bestpractical.com

7

Kevin,

Thanks for the reply. Let me try to me a little more forthcoming on the info.
We have set up our RT session to use LDAP, and to NOT auto create.
We had a ticket where the owner did a “reply” and added an outside email address to the “Cc”.
This ‘outside’ person replied to the ticket and was getting about every hour
on the hours or so.
We bounced the RT session and that stopped it.
From what I understand, our mailserver is ‘polled’ by RT to get any new email sent and pulls
it into RT-mailgate. There it tries to do whatever and is kicked out due to either LDAP or
privileges. All this sounds good so far. From that point, I’m trying to figure out if it is
sending itself a copy of that rejection and causing a loop. We don’t use “ParseCc’s” so we
didn’t set the Regexp in our settings. I’m going to try that and see if that stops the loop
from happening.
That’s all I can figure at the moment.
My questions was oriented around rt-mailgate in hopes of finding out of it had some sort of
re-generative of retry for a number of times or something that I could set to
make it happen only once.

I’m thinking the REGexp setting will resolve the problem.

Kenn

It sounds like you’re using fetchmail, but you don’t say how fetchmail
is feeding mail into rt-mailgate. You really need to dig into your
mail logs and figure out what the error flow is.

-kevin> On Fri, Aug 6, 2010 at 12:36 PM, Kevin Falcone <[1]falcone@bestpractical.com> wrote:

 On Wed, Aug 04, 2010 at 04:10:46PM -0700, Kenneth Crocker wrote:
 > We've had a particular situation come up a couple times now where someone "outside" our
 > approved (LDAP) organization, was added as an 'other' Cc to a reply email from RT. That
 person
 > then tried to do a reply back (into RT) and RT kicked it out, as it should have. However,
 it
 > seems RT is continuing to send the error message over and over (days). Is there a "for how
 > long" or a "X times" setting in rt-mailgate we can modify to stop this endless error
 message?

 Kenn

 You've not really provided a lot to go on. Is the original mail still
 in a postfix queue and being retried because rt-mailgate returns a
 TEMPFAIL error or something else? Without some actual data it is
 hard.
 -kevin

 Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
 Buy a copy at [2]http://rtbook.bestpractical.com

References

Visible links

  1. mailto:falcone@bestpractical.com
  2. http://rtbook.bestpractical.com/

Discover RT’s hidden secrets with RT Essentials from O’Reilly Media.
Buy a copy at http://rtbook.bestpractical.com

8

Kevin,

Yep. I figured that. At least it will get me on the right track.

Thanks.

Kenn
LBNLOn Mon, Aug 9, 2010 at 7:57 AM, Kevin Falcone falcone@bestpractical.comwrote:

On Fri, Aug 06, 2010 at 08:20:25PM -0700, Kenneth Crocker wrote:

Kevin,

Thanks for the reply. Let me try to me a little more forthcoming on
the info.
We have set up our RT session to use LDAP, and to NOT auto create.
We had a ticket where the owner did a “reply” and added an outside
email address to the “Cc”.
This ‘outside’ person replied to the ticket and was getting about every hour
on the hours or so.
We bounced the RT session and that stopped it.
From what I understand, our mailserver is ‘polled’ by RT to get any
new email sent and pulls
it into RT-mailgate. There it tries to do whatever and is kicked out
due to either LDAP or
privileges. All this sounds good so far. From that point, I’m trying
to figure out if it is
sending itself a copy of that rejection and causing a loop. We don’t
use “ParseCc’s” so we
didn’t set the Regexp in our settings. I’m going to try that and see
if that stops the loop
from happening.
That’s all I can figure at the moment.
My questions was oriented around rt-mailgate in hopes of finding out
of it had some sort of
re-generative of retry for a number of times or
something that I could set to
make it happen only once.

I’m thinking the REGexp setting will resolve the problem.

Kenn

It sounds like you’re using fetchmail, but you don’t say how fetchmail
is feeding mail into rt-mailgate. You really need to dig into your
mail logs and figure out what the error flow is.

-kevin

On Fri, Aug 6, 2010 at 12:36 PM, Kevin Falcone <[1] falcone@bestpractical.com> wrote:

 On Wed, Aug 04, 2010 at 04:10:46PM -0700, Kenneth Crocker wrote:
 > We've had a particular situation come up a couple times now where

someone “outside” our

 > approved (LDAP) organization, was added as an 'other' Cc to a

reply email from RT. That

 person
 > then tried to do a reply back (into RT) and RT kicked it out, as

it should have. However,

 it
 > seems RT is continuing to send the error message over and over

(days). Is there a "for how

 > long" or a "X times" setting in rt-mailgate we can modify to stop

this endless error

 message?

 Kenn

 You've not really provided a lot to go on. Is the original mail

still

 in a postfix queue and being retried because rt-mailgate returns a
 TEMPFAIL error or something else? Without some actual data it is
 hard.
 -kevin

 Discover RT's hidden secrets with RT Essentials from O'Reilly Media.
 Buy a copy at [2]http://rtbook.bestpractical.com

References

Visible links

  1. mailto:falcone@bestpractical.com
  2. http://rtbook.bestpractical.com/

Discover RT’s hidden secrets with RT Essentials from O’Reilly Media.
Buy a copy at http://rtbook.bestpractical.com

Discover RT’s hidden secrets with RT Essentials from O’Reilly Media.
Buy a copy at http://rtbook.bestpractical.com