Page 1 of 2
Problem with seding mailing
Posted: Fri Oct 19, 2007 10:12 am
by RafalMW
Hello,
I have installed opencms and everything works perfectly, but ...
when I create mailing and send it then it doesn't send any emails.
If I click "Send test mails to the administrators" or "Send test mails to the test recipients" then it works !
But when I start mailing then it doesn't send anything.
On screen there is: Generated E-Mails: 3
Sent E-Mails: 0
I coudn't find any errors in logs.
I know that description of my problem is not enough. I'm new in openemm so I have no idea what could it be.
Maybe there were similar problems and there is "standard" solution ?
Do you have any ideas ?
with best regards,
Rafal
Posted: Tue Oct 23, 2007 2:00 pm
by eleassarathelas
You should check whether you can relay from the machine (I suggest you do test telnet on port 25), that as a first step towards sending emails to check envoys and bouncing again must give permission to log on the following lines --- File: / etc/cron.weekly/sysklogd---
Change the line:
Savelog adm-m 640-g-c-u root $ 4 LOG> / dev / null
To:
Savelog adm-m 644-g-c-u root $ 4 LOG> / dev / null
--- File: / etc/cron.weekly/sysklogd---
With that should suffice
Atte.
Peter Santis
Santiago de Chile
Posted: Fri Dec 21, 2007 5:11 pm
by m0rpheu5
i´m with the same problem, did you fix the problem, what could be?
same problem
Posted: Tue Jan 15, 2008 2:28 pm
by petz
Hi,
I am having the exact same Problem here with both:
OpenEMM-5.3.1c-1.i386.rpm
and
OpenEMM-5.3.2-2.i386.rpm
The difference is, under 5.3.1c sending mails actually works!
With 5.3.2-2 sending mails doesn't work!
I am running OpenEMM on CentOS 5.
Error in maillog
Jan 15 14:09:36 foobar sendmail[5271]: unable to write pid to /home/openemm/var/run/sendmail-openemm-queue.pid: Permission denied
The joke is under $BASE/var/run/ reside only these pid-file while running openemm:
Code: Select all
[root@foobar run]# ls -la
total 24
drwxr-xr-x 2 openemm openemm 4096 Jan 15 14:09 .
drwxr-xr-x 6 openemm openemm 4096 Jan 15 14:08 ..
srwxrwxr-x 1 openemm openemm 0 Jan 15 14:09 bav.sock
-rw-rw-r-- 1 openemm openemm 10 Jan 15 14:09 bav-update.log
-rw-rw-r-- 1 openemm openemm 5 Jan 15 14:09 openemm.pid
-rw-rw-r-- 1 openemm openemm 5 Jan 15 14:09 redirection.pid
-rw-rw-r-- 1 openemm openemm 23 Jan 15 14:17 slrtscan.save
The Script that actually spawns the sendmail Queues should normally create the following pidfiles, but idt doesn't!!!
sendmail-openemm-admin.pid
sendmail-openemm-queue.pid
sendmail-${cq}.pid
mailer.sh
Code: Select all
$sm -q5m -bd
for cq in clientqueue mqueue-client; do
if [ -d /var/spool/$cq ]; then
mproceed "client queue $cq"
$sm -q5m -OQueueDirectory=/var/spool/$cq -OPidFile=$BASE/var/run/sendmail-${cq}.pid
fi
done
mproceed "admin queue"
$sm -q1m -NNEVER -OQueueDirectory=$BASE/var/spool/ADMIN -OPidFile=$BASE/var/run/sendmail-openemm-admin.pid
mproceed "mail queue"
$sm -q1m -NNEVER -OQueueDirectory=$BASE/var/spool/QUEUE -OPidFile=$BASE/var/run/sendmail-openemm-queue.pid
mend "done"
;;
The Processes are running and everything works.
But ist is a bug, that the pid-files are not created.
openemm 5235 0.0 0.6 85652 6672 pts/0 S 14:09 0:00 python /home/openemm/bin/scripts/pickdist.py
root 5265 0.0 0.2 60648 2532 ? Ss 14:09 0:00 sendmail: accepting connections
root 5266 0.0 0.3 60784 3368 ? S 14:09 0:00 sendmail: ./m0ECO7iY018816 mail.foobar.de.: client EHLO
root 5268 0.0 0.1 49652 1976 ? Ss 14:09 0:00 sendmail: Queue runner@00:01:00 for /home/openemm/var/spool/ADMIN
root 5271 0.0 0.1 49644 1976 ? Ss 14:09 0:00 sendmail: Queue runner@00:01:00 for /home/openemm/var/spool/QUEUE
openemm 5299 0.0 0.5 76340 5440 pts/0 S 14:09 0:00 python /home/openemm/bin/scripts/slrtscn.py
In Version 5.3.2-2 I get the same error, but sendmail doesn't send the mails.
The files under /home/openemm/var/spool/META/ are created, but from there on nothing seems to happen. Under 5.3.1c sending out the mails works.
Please fix it in upcoming versions!
I am Working with Openemm since 5.0, I know it is a bug!
Posted: Fri Jan 18, 2008 4:54 pm
by harfst
-OPidFile=$BASE/var/run/sendmail-openemm-queue.pid
That's a passthru option to sendmail. I found that my version of sendmail doesn't have that option.
I modified the script to do a ps to find the pid after it started and write it out to the file.
Posted: Mon Mar 03, 2008 6:22 pm
by seilemat
Did anybody figure out a solution?
I have exactly the same problem.
I work with OpenSuSE 10.3 (AppArmor is not installed).
Thanks,
Mathias
Posted: Fri Mar 07, 2008 2:14 pm
by chill
I got following error while trying to send mail
Unable to rename temp.file C:\OpenEMM\var\spool\QUEUE\.xmlgen.004436 to qf00000A0000000002A (m), try old fashion link/unlink
Unable to write output for receiver 42 in C:\OpenEMM\var\spool\META\AgnMail=D20080307135156=1=10=3C00000042=liaMngA.xml.gz
...
Atm not 1 mail was send not to admin- or test-user and not to any other user
its the double opt in mail yesterday everithing was working
but after a reboot nothing.
Thanks a lot hope for some easy solution
Posted: Wed Mar 12, 2008 12:44 pm
by seilemat
After I didn't see any chance to solve the problem on the SuSE system, I decided to switch to Ubuntu 7.10-server.
Now I don't have any troubles anymore. Anything works fine
Posted: Wed Mar 12, 2008 2:37 pm
by maschoff
However, it is a pity that OpenEMM causes problems with OpenSuse. Since we only use Red Hat and CentOS it would be great if someone from the Suse camp could offer some insights. (I guess it is a problem with Suse's special sendmail configuration.)
I tried to reproduce this error ...
Posted: Thu Mar 13, 2008 12:36 am
by christian.lang
Following Results with OpenEMM 5.4 on a running OpenSUSE 10 with latest updates. No problems with OpenEMM 5.3.1 before.
According to mailgun and pickdist Logfiles everything is fine:
mailgun:
[13.03.2008 00:23:14] DEBUG/execute/(2/5/97/370): Successful end
pickdist:
[13.03.2008 00:23:39] 4985 INFO/block: Unpacking /home/openemm/var/spool/META/AgnMail=D20080313002314=2=97=002=liaMngA.xml.gz
The Mailing was sent after some minutes (and arrived) and the QUEUE was empty. BUT: Status Information in Send Mailing still says that 6 were produced but 0 sent.
I wonder if it could have something to do with
openemm 5049 84.0 0.4 8464 4764 pts/1 R 00:07 18:39 python /home/openemm/bin/scripts/slrtscn.py
which consumes all available CPU ?
...
Posted: Thu Mar 13, 2008 12:48 am
by christian.lang
I rolled my Installation back to the 5.3.1 Source for the moment...
If i can somehow help to find out hints to resolve this issue please let me know - unfortunately i don't have too much know-how regarding python ...
Posted: Thu Mar 13, 2008 9:33 am
by maschoff
slrtscn.py is our "syslog realtime scanner" which analyzes log files for sent mails and bounces and writes information to files to be processed by update.py. If slrtscan.py is very busy this could mean that is has to handle a lot of bounces.
Posted: Thu Mar 13, 2008 9:49 am
by christian.lang
ma wrote:slrtscn.py is our "syslog realtime scanner" which analyzes log files for sent mails and bounces and writes information to files to be processed by update.py. If slrtscan.py is very busy this could mean that is has to handle a lot of bounces.
There were no bounces at all in that moment - so slrtscan has nothing to do with updating the status of the mailing?
What should i try to find out what the reason is? How does OpenEMM find out, how many mails are sent?
...
Posted: Thu Mar 13, 2008 12:45 pm
by christian.lang
Hello,
I have switched once more to 5.4.0 and tried the same again.
slrtscan now seams to work properly, i cleaned up the maillog File - maybe the script parsed it again when i tried yesterday.
OpenEMM now also shows me that 6 Mails were generated and sent - so now it seems to work properly.
I do not know what the reason really was... the point is that i now started with an empty Maillog File whereas yesterday only all OpenEMM Mailinglogs were cleared because of the fresh OpenEMM Installation.
I think that OpenEMM reparsed the maillog File for some reason which caused the high cpu Usage. Could that be possible?
best regards
Christian
Posted: Thu Mar 13, 2008 12:52 pm
by maschoff
Our specialist (ud) for this script is on vacation until Eastern. Since this seems to be no case of urgency I will not call him back to the office.