N
Naveen
Hi,
I've just laid my hands on doing different kinds of sendmail analysis
with perl. One of the first things that i require to do is to have a
custom <what=value> pair logged to the sendmail log file. Is it
possible ? What do i need to add to my current mailing code to achieve
it --
open MAIL, "|/usr/sbin/sendmail -t -f$redir -OErrorMode=q";
print MAIL "$data";
close MAIL;
Otherwise, can i get sendmail to return the message id (which is
present there for every log entry in the log file) after every entry,
which i can map with the unique id which is internal to my project (and
which i'm trying to log to the sendmail log file)?
Secondly, is anyone aware of an efficient SendMail Log analysis regex
or module which just breaks up all the fields of the logfiles without
any exclusion or unwanted inclusion. What I mean is that the code
should be able to cope with the different error messages and different
status in the files. I've been able to write a program which does the
work almost perfectly, but i still need that 1% error margin to be
eliminated.
P.S - Probably Syslogscan module is not what i am looking for.
Thanks a lot.
Naveen
I've just laid my hands on doing different kinds of sendmail analysis
with perl. One of the first things that i require to do is to have a
custom <what=value> pair logged to the sendmail log file. Is it
possible ? What do i need to add to my current mailing code to achieve
it --
open MAIL, "|/usr/sbin/sendmail -t -f$redir -OErrorMode=q";
print MAIL "$data";
close MAIL;
Otherwise, can i get sendmail to return the message id (which is
present there for every log entry in the log file) after every entry,
which i can map with the unique id which is internal to my project (and
which i'm trying to log to the sendmail log file)?
Secondly, is anyone aware of an efficient SendMail Log analysis regex
or module which just breaks up all the fields of the logfiles without
any exclusion or unwanted inclusion. What I mean is that the code
should be able to cope with the different error messages and different
status in the files. I've been able to write a program which does the
work almost perfectly, but i still need that 1% error margin to be
eliminated.
P.S - Probably Syslogscan module is not what i am looking for.
Thanks a lot.
Naveen