aboutsummaryrefslogtreecommitdiffstats
path: root/framework/src/suricata/doc/Installation_with_CUDA_on_Scientific_Linux_6.txt
blob: 604ee8bcfc472b4dd6f0014d8e2451aae3357d88 (plain)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
Autogenerated on 2012-11-29
from - https://redmine.openinfosecfoundation.org/projects/suricata/wiki/Installation_with_CUDA_on_Scientific_Linux_6


Installation with CUDA on Scientific Linux 6

Hardware used: HP Proliant G7, 16 cores, 30 GB RAM, NVIDIA CUDA Quadro 4000
graphics card
For setup you need to be root. Enter the following:

  mkdir /root/src
  cd /root/src


Pre installation requirements

Run the following command to ensure that you have everything you need for the
installation:

  yum install mpfr-2.4.1-6.el6.x86_64 cpp-4.4.4-13.el6.x86_64 ppl-0.10.2-
  11.el6.x86_64 \
  cloog-ppl-0.15.7-1.2.el6.x86_64 gcc-4.4.4-13.el6.x86_64 kernel-devel-2.6.32-
  131.2.1.el6.x86_64 \
  pcre-devel-7.8-3.1.el6.x86_64 libpcap-devel-1.0.0-
  6.20091201git117cb5.el6.x86_64 \
  yum-plugin-priorities-1.1.26-11.el6.noarch yum-conf-sl6x-1-1.noarch libyaml-
  0.1.3-1.el6.rf.x86_64 \
  libyaml-devel-0.1.3-1.el6.rf.x86_64 libnet-1.1.2.1-2.2.el6.rf.x86_64 flex-
  2.5.35-8.el6.x86_64 \
  bison-2.4.1-5.el6.x86_64 gcc-c++-4.4.4-13.el6.x86_64


CUDA

Download and install NVIDIA CUDA drivers:

  wget http://us.download.nvidia.com/XFree86/Linux-x86_64/270.41.19/NVIDIA-
  Linux-x86_64-270.41.19.run
  chmod +x NVIDIA-Linux-x86_64-270.41.19.run
  ./NVIDIA-Linux-x86_64-270.41.19.run

You also need to download and install the CUDA toolkit for RHEL6 :

  wget http://developer.download.nvidia.com/compute/cuda/4_0/toolkit/
  cudatoolkit_4.0.17_linux_64_rhel6.0.run
  chmod +x cudatoolkit_4.0.17_linux_64_rhel6.0.run
  ./cudatoolkit_4.0.17_linux_64_rhel6.0.run

Make sure the kernel modules are loaded:

  /sbin/modprobe -r nouveau && /sbin/modprobe nvidia

To ensure the proper NVIDIA CUDA modules get loaded on reboot, add the above
line to your /etc/rc.local file.

Suricata

Download and install Suricata:

  wget http://www.openinfosecfoundation.org/download/suricata-1.1beta2.tar.gz

And unpack it:

  tar -xvzf suricata-1.1beta2.tar.gz

Change to the unpacked directory:

  cd suricata-1.1beta2

Compile and install the engine with CUDA support:

  ./configure --enable-gccprotect --enable-profiling --enable-cuda \
  --with-cuda-includes=/usr/local/cuda/include --with-cuda-libraries=/usr/
  local/cuda/lib64/
  make
  make install


Rules

Read the information in Rule_Management_with_Oinkmaster
Add rules to suricata:

  cd /etc/suricata
  wget https://rules.emergingthreatspro.com/open-nogpl/suricata/
  emerging.rules.tar.gz
  tar -xvzf emerging.rules.tar.gz

Make sure your .yaml file includes the /etc/suricata/rules/emerging-*.rules
files (they may need to be uncommented).
Run Suricata as followed:

  cd /etc/suricata
  /usr/local/bin/suricata -c /etc/suricata/suricata.yaml -i eth0