Prelude-lml sensor
"Matteo Michelini" <[email protected]>
| Newsgroups | gmane.comp.security.ids.prelude.user |
|---|---|
| Message-ID | <[email protected]> |
Hi all,
I'm trying to configure prelude-lml sensor but I'm wondering what's the
meaning of prefix-regex into the prelude-lml.conf file.
I thought that prefix-regex was used to match the first part of the log
trail and then the trails not-filtered by this regex were redirected to
the specific *.rules under ruleset/.
But prefix-regex doesn't filter anything.... So I cannot imagine what's
its purpose.
Let me show an example to explain my doubt.
/var/log/apache2/error_log
[Wed Mar 25 11:27:05 2009] [error] [client 192.168.80.105] File does not
exist: /usr/htdocs
/etc/prelude-lml/prelude-lml.conf
[format=apache]
time-format = "%d/%b/%Y:%H:%M:%S"
prefix-regex = "^(?P<hostname>\S+) \S+ \S+ \[(?P<timestamp>.{20})
[+-].{4}\] "
file = /var/log/apache2/error_log
.
.
.
[Pcre=apache]
Ruleset=/etc/prelude-lml/ruleset/httpd.rules
/etc/prelude-lml/plugin.rules
# source plugin pcre-options regex
/var/log/apache2/error_log Pcre[apache] - *
I thought that with this configuration all the trails in error_log were
discarded because the prefix-regex doesn't match but I can see the
events in the prewikka webinterface.
It works even if I set the error_log file under format=syslog into the
prelude-lml.conf
On the other hand, the log format matches the ruleset/httpd.rules ....
So is the filtering only made by the *.rules files?
What's the purpose of prefix-regex and time-format into the
prelude-lml.conf?
Thanks
---
Matteo Michelini
Security Competence Center
Lutech - http://www.lutech.it
Cologno Monzese (MI)
Via W.A.Mozart, 47 - 20093
_______________________________________________
Prelude-user site list
[email protected]
http://lists.prelude-ids.org/mailman/listinfo/prelude-user