Snort mailing list archives
Testing problem, slow
From: "Bryan Taylor" <bcptaylor () comcast net>
Date: Mon, 24 Nov 2003 07:13:13 -0500
I am attempting to conduct basic tests on Snort 2.0.1 (Build 88) running on an RLX server blade with dual 2.8 Xeons and 2G RAM and on a Dell 2650, same setup -- plenty of hardware IMHO to run on a Gig network. The OS is RedHat 9 with kernel 2.4.20. The test consists of throwing artifical traffic to a number of ports on a Cisco switch simultaneously (this is not the issue). I began trying for 40Mbps, and I always get approximately 40% dropped packets. Not believing this, I read about RedHat's libpcap error, and replaced it. I compiled libpcap-current from today from tcpdump.org, dated 22-nov-2003. I compiled snort, being sure to link with the only libpcap on the system (the one I just compiled). I run snort with the default ruleset in a script as such: snort -c /etc/snort/snort.conf -i eth0 -b -Afast -l /var/log/snort/eth0 -I > /var/log/snort/eth0/out 2> /var/log/snort/eth0/err & sleep 2m kill `pidof snort` I have dropped the speed of the traffic to as low as 4Mbps, and I get about the same 40% drop rate. On the 2650, I ran ethereal in place of snort, and ethereal reports very different numbers of packets total and dropped. Both snort and ethereal report unrealistic numbers, such as (ethereal) ~819,000 packet count, ~4100000 dropped. The speed of the traffic coming in from the source is exactly 10,416 packets per second, 64 bytes per packet. In 2 minutes, there ought to be ~1.2 million packets...not far off from the reported. The speed ought to be ~5Mbps, also close to the reported. So what am I missing? Why is it a) so inaccurate in reporting dropped packets or b) so slow? Any input would be greatly appreciated. Taylor bcptaylor () comcast net ------------------------------------------------------- This SF.net email is sponsored by: SF.net Giveback Program. Does SourceForge.net help you be more productive? Does it help you create better code? SHARE THE LOVE, and help us help YOU! Click Here: http://sourceforge.net/donate/ _______________________________________________ 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:
- Testing problem, slow Bryan Taylor (Nov 24)