Snort mailing list archives
Re: Help tuning snort for performance.
From: Joel Esler <jesler () sourcefire com>
Date: Thu, 11 Feb 2010 14:16:00 -0500
Let us know. It probably won't be the last step we'll need to check. Joel On Feb 11, 2010, at 2:07 PM, Andy Berryman wrote:
Can't believe that wasn't the first thing I checked. :slaps forehead: I was so convinced it was a snort issue. I'll report back soon. Thanks, Andy From: Joel Esler [mailto:jesler () sourcefire com] Sent: Thursday, February 11, 2010 1:04 PM To: Andy Berryman Cc: snort-users () lists sourceforge net List Subject: Re: [Snort-users] Help tuning snort for performance. Sure does. That's the one thing I always check for. Snort treats the "flow" as two separate flows. Analyzes both. See if you can eliminate some of that, and your drop packet rate will probably go down. J On Feb 11, 2010, at 2:01 PM, Andy Berryman wrote: This looks like I'm seeing duplicate data doesn't it? tcpdump -i eth1 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 1460:2920(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 1460:2920(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 2920:4380(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 2920:4380(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 4380:5840(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 4380:5840(1460) ack 1 win 64036 18:17:51.270340 IP 172.16.20.19 > 10.42.128.37: gre-proto-0x883e 18:17:51.270340 IP 172.16.20.19 > 10.42.128.37: gre-proto-0x883e 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 5840:7300(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 5840:7300(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 7300:8760(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 7300:8760(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 8760:10220(1460) ack 1 win 64036 18:17:51.270340 IP 10.153.21.99.4239 > 10.153.17.30.445: . 8760:10220(1460) ack 1 win 64036 Thanks, Andy From: Joel Esler [mailto:jesler () sourcefire com] Sent: Thursday, February 11, 2010 12:16 PM To: Andy Berryman Cc: snort-users () lists sourceforge net Subject: Re: [Snort-users] Help tuning snort for performance. Okay, let me tell you what I see when I look at these stats. (BTW -- for those of you reading this list, this exactly the information we need when you write in asking "OMG, I am dropping teh pakets!! OMG") Feb 11 17:30:11 (none) snort[21463]: PatMatch: 82.003% You have a lot of rules running. Feb 11 17:30:11 (none) snort[21463]: Syns/Sec : 123.311 Feb 11 17:30:11 (none) snort[21463]: Syn-Acks/Sec : 125.027 Okay, that's better. Feb 11 17:30:11 (none) snort[21463]: CPU Usage: 85.559% (user) 14.240% (sys) 0.201% (idle) Your box is working really hard. Feb 11 17:30:11 (none) snort[21463]: Max Cached Sessions : 585415 Feb 11 17:30:11 (none) snort[21463]: Max Sessions (interval): 585415 Looks like your session table is full. I usually see this result from two things: 1) Too much traffic going through too small of a box 2) What i like to call "Duplicate packets". (More than one copy of the same traffic being spanned to the same box from two (three, four, five) different spans) Most of the time it's #2. So check your packet dumps and make sure you aren't getting more than one copy of your traffic. Feb 11 17:30:11 (none) snort[21463]: Frag Creates()s/Sec : 43.268 You have a lot of fragmented traffic. Might want to troubleshoot this if possible. Feb 11 17:30:11 (none) snort[21463]: Bytes[60] 24.34% You have lots of small packets. (Fragments? DNS? Encrypted traffic?) I see a bunch of other small indicators, but the above should give you enough to work on. J On Feb 11, 2010, at 12:51 PM, Andy Berryman wrote: Here's the same box, nothing changed. You can see it's even the same snort process running. I'm in the process of trying to get the customer to tune their rules. Trying to make it as least invasive as possible. Feb 11 17:30:11 (none) snort[21463]: Snort Realtime Performance : Thu Feb 11 17:30:11 2010 -------------------------- Feb 11 17:30:11 (none) snort[21463]: Pkts Recv: 3773794 Feb 11 17:30:11 (none) snort[21463]: Pkts Drop: 2583331 Feb 11 17:30:11 (none) snort[21463]: % Dropped: 68.454% Feb 11 17:30:11 (none) snort[21463]: Blocked: 0 Feb 11 17:30:11 (none) snort[21463]: Pkts Filtered TCP: 0 Feb 11 17:30:11 (none) snort[21463]: Pkts Filtered UDP: 0 Feb 11 17:30:11 (none) snort[21463]: Mbits/Sec: 165.153 (wire) Feb 11 17:30:11 (none) snort[21463]: Mbits/Sec: 0.575 (ip fragmented) Feb 11 17:30:11 (none) snort[21463]: Mbits/Sec: 0.344 (ip reassembled) Feb 11 17:30:11 (none) snort[21463]: Mbits/Sec: 2.654 (tcp rebuilt) Feb 11 17:30:11 (none) snort[21463]: Mbits/Sec: 168.149 (app layer) Feb 11 17:30:11 (none) snort[21463]: Bytes/Pkt: 515 (wire) Feb 11 17:30:11 (none) snort[21463]: Bytes/Pkt: 659 (ip fragmented) Feb 11 17:30:11 (none) snort[21463]: Bytes/Pkt: 1549 (ip reassembled) Feb 11 17:30:11 (none) snort[21463]: Bytes/Pkt: 528 (tcp rebuilt) Feb 11 17:30:11 (none) snort[21463]: Bytes/Pkt: 516 (app layer) Feb 11 17:30:11 (none) snort[21463]: KPkts/Sec: 40.054 (wire) Feb 11 17:30:11 (none) snort[21463]: KPkts/Sec: 0.109 (ip fragmented) Feb 11 17:30:11 (none) snort[21463]: KPkts/Sec: 0.028 (ip reassembled) Feb 11 17:30:11 (none) snort[21463]: KPkts/Sec: 0.627 (tcp rebuilt) Feb 11 17:30:11 (none) snort[21463]: KPkts/Sec: 40.707 (app layer) Feb 11 17:30:11 (none) snort[21463]: PatMatch: 82.003% Feb 11 17:30:11 (none) snort[21463]: CPU Usage: 85.559% (user) 14.240% (sys) 0.201% (idle) Feb 11 17:30:11 (none) snort[21463]: Alerts/Sec : 30.315 Feb 11 17:30:11 (none) snort[21463]: Syns/Sec : 123.311 Feb 11 17:30:11 (none) snort[21463]: Syn-Acks/Sec : 125.027 Feb 11 17:30:11 (none) snort[21463]: New Cached Sessions/Sec: 207.727 Feb 11 17:30:11 (none) snort[21463]: Midstream Sessions/Sec : 119.475 Feb 11 17:30:11 (none) snort[21463]: Cached Sessions Del/Sec: 209.275 Feb 11 17:30:11 (none) snort[21463]: Closed Sessions/Sec : 9.421 Feb 11 17:30:11 (none) snort[21463]: TimedOut Sessions/Sec : 255.874 Feb 11 17:30:11 (none) snort[21463]: Pruned Sessions/Sec : 0.000 Feb 11 17:30:11 (none) snort[21463]: Dropped Async Ssns/Sec : 0.000 Feb 11 17:30:11 (none) snort[21463]: Current Cached Sessions: 584948 Feb 11 17:30:11 (none) snort[21463]: Sessions Initializing : 110727 Feb 11 17:30:11 (none) snort[21463]: Sessions Established : 239955 Feb 11 17:30:11 (none) snort[21463]: Sessions Closing : 234445 Feb 11 17:30:11 (none) snort[21463]: Max Cached Sessions : 585415 Feb 11 17:30:11 (none) snort[21463]: Max Sessions (interval): 585415 Feb 11 17:30:11 (none) snort[21463]: Stream Flushes/Sec : 627.252 Feb 11 17:30:11 (none) snort[21463]: Stream Cache Faults/Sec: 765 Feb 11 17:30:11 (none) snort[21463]: Stream Cache Timeouts : 7605 Feb 11 17:30:11 (none) snort[21463]: Frag Creates()s/Sec : 43.268 Feb 11 17:30:11 (none) snort[21463]: Frag Completes()s/Sec : 27.825 Feb 11 17:30:11 (none) snort[21463]: Frag Inserts()s/Sec : 65.710 Feb 11 17:30:11 (none) snort[21463]: Frag Deletes/Sec : 43.302 Feb 11 17:30:11 (none) snort[21463]: Frag AutoFrees/Sec : 15.477 Feb 11 17:30:11 (none) snort[21463]: Frag Flushes/Sec : 27.791 Feb 11 17:30:11 (none) snort[21463]: Current Cached Frags : 64793 Feb 11 17:30:11 (none) snort[21463]: Max Cached Frags : 64794 Feb 11 17:30:11 (none) snort[21463]: Frag Timeouts : 189 Feb 11 17:30:11 (none) snort[21463]: Frag Faults : 0 Feb 11 17:30:11 (none) snort[21463]: New Cached UDP Ssns/Sec: 0.000 Feb 11 17:30:11 (none) snort[21463]: Cached UDP Ssns Del/Sec: 0.000 Feb 11 17:30:11 (none) snort[21463]: Current Cached UDP Ssns: 0 Feb 11 17:30:11 (none) snort[21463]: Max Cached UDP Ssns : 0 Feb 11 17:30:11 (none) snort[21463]: Snort Maximum Performance Feb 11 17:30:11 (none) snort[21463]: ------------------------- Feb 11 17:30:11 (none) snort[21463]: Mbits/Second Feb 11 17:30:11 (none) snort[21463]: ---------------- Feb 11 17:30:11 (none) snort[21463]: Snort: 196.530 Feb 11 17:30:11 (none) snort[21463]: Sniffing: 1180.850 Feb 11 17:30:11 (none) snort[21463]: Combined: 168.488 Feb 11 17:30:11 (none) snort[21463]: uSeconds/Pkt Feb 11 17:30:11 (none) snort[21463]: ---------------- Feb 11 17:30:11 (none) snort[21463]: Snort: 21.018 Feb 11 17:30:11 (none) snort[21463]: Sniffing: 3.498 Feb 11 17:30:11 (none) snort[21463]: Combined: 24.516 Feb 11 17:30:11 (none) snort[21463]: KPkts/Second Feb 11 17:30:11 (none) snort[21463]: ------------------ Feb 11 17:30:11 (none) snort[21463]: Snort: 47.578 Feb 11 17:30:11 (none) snort[21463]: Sniffing: 285.870 Feb 11 17:30:11 (none) snort[21463]: Combined: 40.789 Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Protocol Byte Flows - %Total Flow Feb 11 17:30:11 (none) snort[21463]: -------------------------------------- Feb 11 17:30:11 (none) snort[21463]: TCP: 85.96% Feb 11 17:30:11 (none) snort[21463]: UDP: 0.67% Feb 11 17:30:11 (none) snort[21463]: ICMP: 0.05% Feb 11 17:30:11 (none) snort[21463]: OTHER: 13.32% Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: PacketLen - %TotalPackets Feb 11 17:30:11 (none) snort[21463]: ------------------------- Feb 11 17:30:11 (none) snort[21463]: Bytes[60] 24.34% Feb 11 17:30:11 (none) snort[21463]: Bytes[62] 0.57% Feb 11 17:30:11 (none) snort[21463]: Bytes[63] 0.18% Feb 11 17:30:11 (none) snort[21463]: Bytes[64] 0.41% Feb 11 17:30:11 (none) snort[21463]: Bytes[65] 0.17% Feb 11 17:30:11 (none) snort[21463]: Bytes[66] 0.82% Feb 11 17:30:11 (none) snort[21463]: Bytes[71] 0.96% Feb 11 17:30:11 (none) snort[21463]: Bytes[74] 0.19% Feb 11 17:30:11 (none) snort[21463]: Bytes[76] 0.16% Feb 11 17:30:11 (none) snort[21463]: Bytes[77] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[78] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[80] 0.40% Feb 11 17:30:11 (none) snort[21463]: Bytes[82] 4.56% Feb 11 17:30:11 (none) snort[21463]: Bytes[85] 0.10% Feb 11 17:30:11 (none) snort[21463]: Bytes[86] 0.15% Feb 11 17:30:11 (none) snort[21463]: Bytes[87] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[88] 0.23% Feb 11 17:30:11 (none) snort[21463]: Bytes[90] 0.57% Feb 11 17:30:11 (none) snort[21463]: Bytes[91] 0.27% Feb 11 17:30:11 (none) snort[21463]: Bytes[92] 0.21% Feb 11 17:30:11 (none) snort[21463]: Bytes[93] 0.84% Feb 11 17:30:11 (none) snort[21463]: Bytes[94] 3.95% Feb 11 17:30:11 (none) snort[21463]: Bytes[95] 0.15% Feb 11 17:30:11 (none) snort[21463]: Bytes[97] 0.16% Feb 11 17:30:11 (none) snort[21463]: Bytes[98] 0.18% Feb 11 17:30:11 (none) snort[21463]: Bytes[99] 0.50% Feb 11 17:30:11 (none) snort[21463]: Bytes[102] 0.37% Feb 11 17:30:11 (none) snort[21463]: Bytes[104] 0.58% Feb 11 17:30:11 (none) snort[21463]: Bytes[105] 0.61% Feb 11 17:30:11 (none) snort[21463]: Bytes[106] 0.35% Feb 11 17:30:11 (none) snort[21463]: Bytes[107] 0.20% Feb 11 17:30:11 (none) snort[21463]: Bytes[108] 0.14% Feb 11 17:30:11 (none) snort[21463]: Bytes[109] 1.38% Feb 11 17:30:11 (none) snort[21463]: Bytes[110] 0.20% Feb 11 17:30:11 (none) snort[21463]: Bytes[111] 0.35% Feb 11 17:30:11 (none) snort[21463]: Bytes[113] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[114] 0.26% Feb 11 17:30:11 (none) snort[21463]: Bytes[115] 0.15% Feb 11 17:30:11 (none) snort[21463]: Bytes[116] 0.23% Feb 11 17:30:11 (none) snort[21463]: Bytes[117] 1.02% Feb 11 17:30:11 (none) snort[21463]: Bytes[118] 0.33% Feb 11 17:30:11 (none) snort[21463]: Bytes[119] 0.19% Feb 11 17:30:11 (none) snort[21463]: Bytes[122] 0.49% Feb 11 17:30:11 (none) snort[21463]: Bytes[124] 0.20% Feb 11 17:30:11 (none) snort[21463]: Bytes[126] 0.30% Feb 11 17:30:11 (none) snort[21463]: Bytes[128] 0.13% Feb 11 17:30:11 (none) snort[21463]: Bytes[130] 1.07% Feb 11 17:30:11 (none) snort[21463]: Bytes[134] 0.17% Feb 11 17:30:11 (none) snort[21463]: Bytes[140] 0.14% Feb 11 17:30:11 (none) snort[21463]: Bytes[142] 1.38% Feb 11 17:30:11 (none) snort[21463]: Bytes[145] 0.12% Feb 11 17:30:11 (none) snort[21463]: Bytes[146] 0.19% Feb 11 17:30:11 (none) snort[21463]: Bytes[150] 0.14% Feb 11 17:30:11 (none) snort[21463]: Bytes[154] 0.57% Feb 11 17:30:11 (none) snort[21463]: Bytes[156] 0.13% Feb 11 17:30:11 (none) snort[21463]: Bytes[158] 2.91% Feb 11 17:30:11 (none) snort[21463]: Bytes[162] 1.51% Feb 11 17:30:11 (none) snort[21463]: Bytes[164] 0.23% Feb 11 17:30:11 (none) snort[21463]: Bytes[166] 0.24% Feb 11 17:30:11 (none) snort[21463]: Bytes[168] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[170] 0.72% Feb 11 17:30:11 (none) snort[21463]: Bytes[172] 0.36% Feb 11 17:30:11 (none) snort[21463]: Bytes[174] 0.29% Feb 11 17:30:11 (none) snort[21463]: Bytes[178] 0.25% Feb 11 17:30:11 (none) snort[21463]: Bytes[182] 0.28% Feb 11 17:30:11 (none) snort[21463]: Bytes[186] 0.53% Feb 11 17:30:11 (none) snort[21463]: Bytes[188] 0.51% Feb 11 17:30:11 (none) snort[21463]: Bytes[190] 0.12% Feb 11 17:30:11 (none) snort[21463]: Bytes[193] 0.23% Feb 11 17:30:11 (none) snort[21463]: Bytes[194] 0.37% Feb 11 17:30:11 (none) snort[21463]: Bytes[196] 0.24% Feb 11 17:30:11 (none) snort[21463]: Bytes[198] 0.24% Feb 11 17:30:11 (none) snort[21463]: Bytes[202] 0.43% Feb 11 17:30:11 (none) snort[21463]: Bytes[206] 0.13% Feb 11 17:30:11 (none) snort[21463]: Bytes[208] 0.10% Feb 11 17:30:11 (none) snort[21463]: Bytes[210] 0.15% Feb 11 17:30:11 (none) snort[21463]: Bytes[214] 0.25% Feb 11 17:30:11 (none) snort[21463]: Bytes[218] 0.13% Feb 11 17:30:11 (none) snort[21463]: Bytes[222] 0.16% Feb 11 17:30:11 (none) snort[21463]: Bytes[228] 0.14% Feb 11 17:30:11 (none) snort[21463]: Bytes[230] 0.70% Feb 11 17:30:11 (none) snort[21463]: Bytes[234] 0.15% Feb 11 17:30:11 (none) snort[21463]: Bytes[238] 0.37% Feb 11 17:30:11 (none) snort[21463]: Bytes[242] 0.41% Feb 11 17:30:11 (none) snort[21463]: Bytes[246] 0.35% Feb 11 17:30:11 (none) snort[21463]: Bytes[250] 0.10% Feb 11 17:30:11 (none) snort[21463]: Bytes[254] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[262] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[330] 0.10% Feb 11 17:30:11 (none) snort[21463]: Bytes[441] 0.11% Feb 11 17:30:11 (none) snort[21463]: Bytes[1230] 0.56% Feb 11 17:30:11 (none) snort[21463]: Bytes[1414] 0.14% Feb 11 17:30:11 (none) snort[21463]: Bytes[1442] 0.27% Feb 11 17:30:11 (none) snort[21463]: Bytes[1474] 1.59% Feb 11 17:30:11 (none) snort[21463]: Bytes[1486] 0.94% Feb 11 17:30:11 (none) snort[21463]: Bytes[1506] 0.36% Feb 11 17:30:11 (none) snort[21463]: Bytes[1514] 23.30% Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: TCP Port Flows Feb 11 17:30:11 (none) snort[21463]: -------------- Feb 11 17:30:11 (none) snort[21463]: Port[25] 1.54% of Total, Src: 6.84% Dst: 93.16% Feb 11 17:30:11 (none) snort[21463]: Port[80] 11.97% of Total, Src: 88.38% Dst: 11.62% Feb 11 17:30:11 (none) snort[21463]: Port[135] 0.33% of Total, Src: 45.33% Dst: 54.67% Feb 11 17:30:11 (none) snort[21463]: Port[139] 0.36% of Total, Src: 47.18% Dst: 52.82% Feb 11 17:30:11 (none) snort[21463]: Port[389] 0.42% of Total, Src: 76.60% Dst: 23.40% Feb 11 17:30:11 (none) snort[21463]: Port[443] 1.33% of Total, Src: 86.63% Dst: 13.37% Feb 11 17:30:11 (none) snort[21463]: Port[445] 44.86% of Total, Src: 46.94% Dst: 53.06% Feb 11 17:30:11 (none) snort[21463]: Port[515] 0.22% of Total, Src: 7.96% Dst: 92.04% Feb 11 17:30:11 (none) snort[21463]: Ports[High<->High]: 38.94% Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: UDP Port Flows Feb 11 17:30:11 (none) snort[21463]: -------------- Feb 11 17:30:11 (none) snort[21463]: Port[53] 5.34% of Total, Src: 68.10% Dst: 31.90% Feb 11 17:30:11 (none) snort[21463]: Port[67] 0.14% of Total, Src: 46.95% Dst: 53.05% Feb 11 17:30:11 (none) snort[21463]: Port[88] 3.46% of Total, Src: 62.28% Dst: 37.72% Feb 11 17:30:11 (none) snort[21463]: Port[123] 0.30% of Total, Src: 50.00% Dst: 50.00% Feb 11 17:30:11 (none) snort[21463]: Port[137] 6.93% of Total, Src: 51.04% Dst: 48.96% Feb 11 17:30:11 (none) snort[21463]: Port[138] 0.73% of Total, Src: 50.00% Dst: 50.00% Feb 11 17:30:11 (none) snort[21463]: Port[161] 10.38% of Total, Src: 43.65% Dst: 56.35% Feb 11 17:30:11 (none) snort[21463]: Port[389] 0.68% of Total, Src: 42.25% Dst: 57.75% Feb 11 17:30:11 (none) snort[21463]: Port[514] 2.58% of Total, Src: 46.69% Dst: 53.31% Feb 11 17:30:11 (none) snort[21463]: Port[902] 1.11% of Total, Src: 0.00% Dst: 100.00% Feb 11 17:30:11 (none) snort[21463]: Ports[High<->High]: 73.35% Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: ICMP Type Flows Feb 11 17:30:11 (none) snort[21463]: --------------- Feb 11 17:30:11 (none) snort[21463]: Type[0] 27.10% of Total Feb 11 17:30:11 (none) snort[21463]: Type[3] 41.30% of Total Feb 11 17:30:11 (none) snort[21463]: Type[8] 31.50% of Total Feb 11 17:30:11 (none) snort[21463]: Type[11] 0.10% of Total Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Feb 11 17:30:11 (none) snort[21463]: Snort Setwise Event Stats Feb 11 17:30:11 (none) snort[21463]: ------------------------- Feb 11 17:30:11 (none) snort[21463]: Total Events: 8303325 Feb 11 17:30:11 (none) snort[21463]: Qualified Events: 203 Feb 11 17:30:11 (none) snort[21463]: Non-Qualified Events: 8303122 Feb 11 17:30:11 (none) snort[21463]: %Qualified Events: 0.0024% Feb 11 17:30:11 (none) snort[21463]: %Non-Qualified Events: 99.9976% From: Joel Esler [mailto:jesler () sourcefire com] Sent: Thursday, February 11, 2010 11:35 AM To: Andy Berryman Cc: snort-users () lists sourceforge net Subject: Re: [Snort-users] Help tuning snort for performance. Frag3 tuning shouldn't affect syn/sec and syn-ack/sec. The stats you posted below tells me two things: 1) Your syn and syn/acks aren't 1:1. 2) Your packet size is small (VPN? GRE? DNS?) J On Feb 11, 2010, at 12:26 PM, Andy Berryman wrote: Actually, it's not. The syn/sec and the syn-ack/sec were really close to 1:1 before I started in on Frag3 tuning. -bash-2.05b# tcpdump -i eth1 17:04:23.835615 IP 172.17.23.8.1494 > 10.151.100.3.59782: P 2141564463:2141564471(8) ack 1794773895 win 63861 17:04:23.835615 IP 172.17.23.8.1494 > 10.151.100.3.59782: P 0:8(8) ack 1 win 63861 17:04:23.839616 IP 10.153.13.32.2738 > 10.153.21.43.1433: . ack 4501 win 63836 17:04:23.839616 IP 10.153.13.32.2738 > 10.153.21.43.1433: . ack 4501 win 63836 17:04:23.839616 IP 10.153.19.13.1433 > 10.153.19.12.4744: P 1:134(133) ack 50 win 65485 17:04:23.839616 IP 10.153.19.13.1433 > 10.153.19.12.4744: P 1:134(133) ack 50 win 65485 17:04:23.839616 IP 10.153.13.32.2738 > 10.153.21.43.1433: . ack 4501 win 63836 17:04:23.839616 IP 10.153.13.32.2738 > 10.153.21.43.1433: . ack 4501 win 63836 17:04:23.839616 IP 10.174.3.83.2180 > 10.16.14.14.445: P 63:1239(1176) ack 4537 win 64316 17:04:23.839616 IP 10.174.3.83.2180 > 10.16.14.14.445: P 63:1239(1176) ack 4537 win 64316 17:04:23.839616 IP 10.150.90.25.1205 > 10.153.1.171.1433: P 142:316(174) ack 87 win 63941 17:04:23.839616 IP 10.150.90.25.1205 > 10.153.1.171.1433: P 142:316(174) ack 87 win 63941 17:04:23.839616 IP 172.16.20.19 > 10.42.128.37: gre-proto-0x883e 17:04:23.839616 IP 172.16.20.19 > 10.42.128.37: gre-proto-0x883e 30.10.25.3278: P 312:416(104) ack 293 win 64475 187 packets captured 12341 packets received by filter 11942 packets dropped by kernel From: Joel Esler [mailto:jesler () sourcefire com] Sent: Thursday, February 11, 2010 11:16 AM To: Andy Berryman Cc: snort-users () lists sourceforge net Subject: Re: [Snort-users] Help tuning snort for performance. Is your sensor in front of a firewall (or similar)? It looks like it: Feb 11 16:19:11 (none) snort[21463]: Syns/Sec : 366.021 Feb 11 16:19:11 (none) snort[21463]: Syn-Acks/Sec : 150.862 Joel ------------------------------------------------------------------------------ SOLARIS 10 is the OS for Data Centers - provides features such as DTrace, Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW http://p.sf.net/sfu/solaris-dev2dev_______________________________________________ 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 -- Joel Esler 302-223-5974 This message from Cymtec Systems, Inc. contains confidential information and is solely for the use of the recipient(s) named above. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this message in error and that any review, disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error, please destroy it immediately and notify Cymtec Systems, Inc. by telephone at +1.314.993.8700 or by return e-mail. -- Joel Esler 302-223-5974 This message from Cymtec Systems, Inc. contains confidential information and is solely for the use of the recipient(s) named above. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this message in error and that any review, disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error, please destroy it immediately and notify Cymtec Systems, Inc. by telephone at +1.314.993.8700 or by return e-mail. -- Joel Esler 302-223-5974 This message from Cymtec Systems, Inc. contains confidential information and is solely for the use of the recipient(s) named above. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this message in error and that any review, disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error, please destroy it immediately and notify Cymtec Systems, Inc. by telephone at +1.314.993.8700 or by return e-mail. -- Joel Esler 302-223-5974 This message from Cymtec Systems, Inc. contains confidential information and is solely for the use of the recipient(s) named above. If you are not the intended recipient or an agent responsible for delivering it to the intended recipient, you are hereby notified that you have received this message in error and that any review, disclosure, copying, distribution or use of the contents of this message is strictly prohibited. If you have received this message in error, please destroy it immediately and notify Cymtec Systems, Inc. by telephone at +1.314.993.8700 or by return e-mail.
-- Joel Esler 302-223-5974
------------------------------------------------------------------------------ SOLARIS 10 is the OS for Data Centers - provides features such as DTrace, Predictive Self Healing and Award Winning ZFS. Get Solaris 10 NOW http://p.sf.net/sfu/solaris-dev2dev
_______________________________________________ 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:
- Re: Help tuning snort for performance., (continued)
- Re: Help tuning snort for performance. Alex Kirk (Feb 11)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Alex Kirk (Feb 11)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Joel Esler (Feb 11)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Joel Esler (Feb 11)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Joel Esler (Feb 11)
- Message not available
- Re: Help tuning snort for performance. Joel Esler (Feb 11)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Joel Esler (Feb 11)
- Re: Help tuning snort for performance. Seth Art (Feb 12)
- Re: Help tuning snort for performance. Andy Berryman (Feb 26)
- Re: Help tuning snort for performance. Joel Esler (Feb 26)
- Re: Help tuning snort for performance. Andy Berryman (Feb 11)
- Re: Help tuning snort for performance. Alex Kirk (Feb 11)
- Re: Help tuning snort for performance. Eoin Miller (Feb 11)