Snort mailing list archives

Re: Building snort


From: Jon Urionaguena <juriona () nesys-st com>
Date: Wed, 14 May 2008 16:30:36 +0200

Thanx Todd,

That was all!! I skipped a "config enable" sentence in the config file. 
This happens because of using a sample file and commenting lines...

Anyway, same config file for another snort binnary didn't show up so 
many warnings!! Why could that happen?

The system is a Debian Etch amd64 running on a Dual Core Xeon 1.6 GHz 
(Dell PE2950), 4 GB RAM and an Intel Pro/1000 PT card sniffing all the 
traffic (4 links of 1Gbps spanned (Cisco) to one).

Regards,

Jon

Todd Wease escribió:
Do you have the following option in your snort.conf:

config enable_decode_oversized_alerts

This should be the option that enables that alert.

Still I'm guessing that something else is wrong.  What kind of OS are
you running on?

Jon Urionaguena wrote:
  
Thanx Todd,

The output I´m using is:

output log_unified: filename snort, limit 9000

Which, in my system, logs text to an alert file, and binary format to
snort.log. Both files are growing too fast. The alert one is the one I
can "normally" read (text), that's why I suppose that the origin of this
warning is the one that makes snort log every packet in the unified
format. I can be in a big mistake... I will change the output and have a
look at the logs in a tcpdump format reader (aka wireshark) and give
more feedback.

    
It should read that the IP datagram length is greater than the pcap
      
captured length from the IP header on.
We have the option "config disable_decode_alerts" set...
Could it be an error with the pf_ring and modified libpcap
implementation we are using?

    
Are you specifying a snaplen to snort?
      
No, I'm not. The thing is that a 2.7 binnary works ok (seems to...) with
the same config file and same startup options. That's why I'm supposing
that the error is not in the config, but in the binnaries... Maybe a
compilation option. Don't know any.

Regards,

Jon

Todd Wease escribió:
    
Hello Jon,

This message is actually wrong:

"[**] [116:6:1] (snort_decoder) WARNING: IP dgm len > IP Hdr len! [**]"

It should read that the IP datagram length is greater than the pcap
captured length from the IP header on.

Also, you shouldn't see messages like that in a unified file and I'm not
sure any postprocessor would show the data that way.  Sounds like you're
just looking at a text alert file.

Are you specifying a snaplen to snort?  If so, remove it.  If not, try
logging in tcpdump mode and look at the resulting snort.log.<timestamp>
in Wireshark and see what those packets look like.

Todd

Jon Urionaguena wrote:
 
      
Hi all,

I am building a high speed IDS system trying to use pfring
extensions, with libpcap modified. I'm trying to work with unified
output format.

Kernel is built ok. New libpcap seems ok too.

When I build snort (downloaded 2.7 and 2.8.1),  I try to make it
static building against the libpcap.a just generated. All I can see
is that the resulting binnary does not give any dependence (ldd)
against any libpcap.

So I launch it... But the unified file format it generates is wrong
because it´s full of messages of this kind:

"[**] [116:6:1] (snort_decoder) WARNING: IP dgm len > IP Hdr len! [**]"

Even if we have the option to avoid these messages in snort.conf. I
guess I get a message for each packet we receive... The logs get
enormous (50 Mbps link) and without any value.

Any hint?? Any other data I should supply?

On the other side, I have an old snort binnary linked to the modified
libpcap (that's what ldd says...) that seems to work ok (loads pfring
on startup and gives normal alerts), but I compiled it before we had
the pfring change (kernel and new libpcaps)??? It shouldn't work this
way.

Building snort is being a strange experience for me, because I get to
many issues I can not fully understand... The flags I try to pass to
configure script never seem to do anything... I'm turning crazy.

Thanx in advance,

    
        
  
      


  

-- 

Jon 


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft 
Defy all challenges. Microsoft(R) Visual Studio 2008. 
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Snort-users mailing list
Snort-users () lists sourceforge net
Go to this URL to change user options or unsubscribe:
https://lists.sourceforge.net/lists/listinfo/snort-users
Snort-users list archive:
http://www.geocrawler.com/redir-sf.php3?list=snort-users


Current thread: