tcpdump mailing list archives
questions on -B, performance, mbufs, and libpcap buffers
From: Jon Schipp <jonschipp () gmail com>
Date: Tue, 13 Sep 2011 23:47:17 -0400
Hey guys, I have a few questions, hopefully someone can set me straight. Info: *I'm on FreeBSD 8.2* *out of a couple million packets, in a few hours time, I drop around 4000* First off are all packets stored in mbufs? And if so, is it possible to increase the amount of mbuf clusters to improve tcpdump performance? I looked up how to do it, but wasn't sure if all packets were stored in mbufs, I know that socket connections (socket()) are. When capturing network traffic during peak hours, netstat -m didn't show that I was running out of mbufs, had a few hundred available. Again, questioning myself, I wasn't sure if packets for tcpdump were stored in mbufs, I would expect them to run out if the kernel was dropping them. Though, I'm assuming that all packets are stored in mbufs and that bpf just gets a copy of what's in the mbufs. Correct me if I'm wrong. I'm trying to go through Steven's TCP/IP Vol 2, it's kicking my butt though (not a programmer *maybe someday*). I'm hoping someone can clear some of this up for me. Second, does -B set an application buffer independent of the libpcap buffer? What's the default size? Can capture improvements be made by increasing this value? *I'm dropping packets at the kernel, according to tcpdump* Is the argument for -B in Kilobits? ( if this isn't associated with the kernel/libpcap drops, increasing it wouldn't help, i don't think anyway, still curious though) I see that the libpcap buffer can be increased with sysctl since libpcap initializes its buffer amount with bpf sizes: <majordomo () lists tcpdump org?subject=subscribe%20to%20tcpdump-workers&body=subscribe%20tcpdump-workers>net.bpf.maxbufsize: net.bpf.bufsize: I plan on doing tests with these soon. I'm trying to collect as much information as possible first. Third, when tcpdump reports how many packets are dropped by the kernel is this the same value as the packets dropped by libpcap? In otherwords, does libpcap ask the kernel for the kernel drop amount and report it as "packets dropped by kernel". Or will they have different values? The reason I ask is that the program ntop has a field that shows "dropped by (libpcap)" but not a "dropped by kernel" If they're different, is there a way to find out the root cause: kernel or libpcap Thanks!!! -- - Jon -- ------------------------------------------------------------------ VMB: 812-682-0231 Dubois County Linux User Group - http://www.dclinux.org Southern Indiana Computer Klub - http://sickbits.networklabs.org Bloomington FOOLS - http://www.bloomingtonfools.org/ BloomingLabs - http://www.bloominglabs.org ISSA-Kentuckiana - http://issa-kentuckiana.org GPG Key ID: 810903CB Key fingerprint = 0069 ED69 EABB DF84 5983 AD3C 6C20 BEFD 8109 03CB - This is the tcpdump-workers list. Visit https://cod.sandelman.ca/ to unsubscribe.
Current thread:
- questions on -B, performance, mbufs, and libpcap buffers Jon Schipp (Sep 13)
- Re: questions on -B, performance, mbufs, and libpcap buffers Guy Harris (Sep 13)
- Re: questions on -B, performance, mbufs, and Jon Schipp (Sep 15)
- Re: questions on -B, performance, mbufs, and Guy Harris (Sep 15)
- Re: questions on -B, performance, mbufs, and Jon Schipp (Sep 27)
- Re: questions on -B, performance, mbufs, and Rick Jones (Sep 28)
- Re: questions on -B, performance, mbufs, and Sanjay Sundaresan (Sep 28)
- Re: questions on -B, performance, mbufs, and Rick Jones (Sep 28)
- Re: questions on -B, performance, mbufs, and Jon Schipp (Sep 15)
- Re: questions on -B, performance, mbufs, and Guy Harris (Sep 28)
- Re: questions on -B, performance, mbufs, and libpcap buffers Guy Harris (Sep 13)