Page 1 of 1

Bounce stats problem

Posted: Tue Sep 11, 2007 10:00 pm
by eleassarathelas
hello to all:


my name is Peter santis, I have installed time recently agnitas openemm, on a machine linux (4,0 debian) and works of wonders, nevertheless I have experienced statistical problems with ace, in individual with emails bounced, somebody has lguna idea of so that this happens.


atte.

Peter Santis
Santiago of Chile

bounce overview stats wrong

Posted: Thu Nov 22, 2007 2:16 pm
by olipage
Hi guys,

i discovered similar problems with the statistics on bouncing:

We have successfully implemented openemm within our organisation, but
the statistic tab seem not to work properly. i just wonder, whether
there are already known issues in this field.

two issues from me: statistics on bounces seem not to work properly:
csv sheet reports 150 bounces, system reports about 200 hardbounces,
and additionally 400 softbounces for a total of 608 bounces.

overview pages says: 125 bounces - corresponds to the csv file.
click on bounces reveals following:


Bounces


*Softbounces * *Anzahl*
------------------------------------------------------------------------
Probleme mit der Mailbox 55 Probleme mit dem Mailserver 43
Irreguläre Bounces 50 Sonstige Softbounces 212

*Hardbounces * *Anzahl*
------------------------------------------------------------------------
Unbekannte Adresse 200 Unbekannter Domainname 38 Sonstige
Hardbounces 10
------------------------------------------------------------------------
*Gesamt* *608 *

Empfänger auf Status 'Bounced' gesetzt* 125

bounces should be a sum of 125, not 608 shouldn't it?

export of bounced recipient is 253 recipients? i can't really
reconstruct what is happening here.

any suggestions where to dig? we are running the vmware version, with 5.3 - implmemented jsut a few weeks ago.

kind regards,

Oliver

Posted: Mon Nov 26, 2007 4:04 pm
by maschoff
Could you please elaborate on your statement "csv sheet reports 150 bounces" vs. "overview pages says: 125 bounces - corresponds to the csv file", because 150 <> 125.

At first, you can ignore the soft bounces regarding status "bounced" because they are no hard bounces and, therefore, are not flagged as "bounced".

Hard bounces (248) are almost twice as much as recipients marked as "bounced" (125), so maybe you sent two mailings in parallel or almost at the same time to the same addresses?

Posted: Fri Dec 07, 2007 9:36 am
by olipage
hi,

in fact you are right. i did send out 2 mailings, unfortunately the second mailing included mailadresses from the first one.

i did not think about that, maybe this is it. thanks a lot in the meantime, have to reflect on this.

kind regards,

Oliver

export of bounces breaks up csv on neutral gender

Posted: Fri Dec 07, 2007 9:50 am
by olipage
hi,

just working on the stat thing. i found out, that the export of bounces (hardbounces runterladen) break the csv scheme.

i am running your vmware appliance, and if one of the recipients is neither male no female, the csv file does not make a linebreak before that account.

in the csv file it looks for example like this:

"Herr";"Benjamin";"Bartlett";"something@something.at"";"Albert & Antonia";"Bauman";"something2@something.at"

any ideas about that?

kind regards,

Oliver

Posted: Fri Jan 25, 2008 4:52 pm
by maschoff
olipage, your bug posting was confirmed by us (see http://sourceforge.net/tracker/index.ph ... tid=848488)
and will be fixed in release 5.4. Thank you for your report!