Go Premium for a chance to win a PS4. Enter to Win

x
  • Status: Solved
  • Priority: Medium
  • Security: Public
  • Views: 407
  • Last Modified:

Network IDS

Why, how and where do you decide the best location to place IDS sensors on a network and how do they connect back to the IDS.
0
SydNal2009
Asked:
SydNal2009
  • 3
  • 2
4 Solutions
 
btanExec ConsultantCommented:
Do take some time to analyse the topology of your network to understand the paths that an attacker can use to gain access to your network. Once you identified those potential entry/exit points for the deployed critical service to deter and detect timely the many targeted attacks against your network, you need then consider if the those sensors on your network to watch for potential hostile activity. Most of the time, it is looking out for intrusions, anomalies and illegitimate activities at those entry/exit boundaries on your network.

I do not want to be too prescriptive nor go into details of IDS capability and placement in zones as environment varied. Rightfully segment segregated (e.g. Perimeter protection, Extranets,Remote access, Intranets etc) will each have such sensor whether inline (active) or passive (transparent) - to these, it can still listen and block (send RST etc) if needed. The SSL inspection (opening the encrypted package required more intimate involvement and needed to be inline (some say termination simulating as server  using new private key and some go into Man in the middle type as transparent to import the actual server private key) but commonly.. it can get complicated..

But I do see the main question is potentially should IDS/IPS be in front or behind the FW. Of course if the FW is a next generation which is modular and embedded with such capability in same h/w this is in a way fulfilling the checks needed. This looks like the way forward for those to ease operational and maintenance (really depends on availability considerations and the many segments to monitor).

See "IDS Placement Strategies" in
http://www.sans.org/security-resources/idfaq/ids_redun.php

However, it traditionally are two separate boxes. I rephrase it as how much you want the sensor or probe to (or can) see all the traffic.

> in front of FW, see everything and provide really wider situational awareness but very noisy and can demand high throughput

> behind FW, can only see filtered (supposedly) traffic and flag as per inspection. You only can check what you can see. Serve with optimal throughput and more targeted and attempt to identify those trusted but anomalous that evaded FW (actually if port 80 or 443, they tend to go through) based on port no. if FW is service/app aware then it is another layer of check if traffic pass FW inspection.

Maybe i just summarise the above 2 pt as put it between any two network zones that have differing trust levels. hope it helps with this mind of its own
0
 
Rich RumbleSecurity SamuraiCommented:
You basically want them INSIDE your network, looking at what traffic makes it in and what is going out. Do not put IDS's outside, looking at everything that is coming to the network, because the firewalls/routers are going to drop some of that, and it's too many false positives to look into.
As to sensors, you put them where you can basically, if you can use one sensor and it can keep up with the traffic, do that. It's typically a physical issue where you put them more than it's them not being able to handle the amount of traffic. Put one in each remote office for instance, because you can't see that traffic any other way.
-rich
0
 
btanExec ConsultantCommented:
Do also consider not have it behind content filter as the explanation of not seeing the filtered applies too :) may also need to consider the snmp and siems network collection if the log segment. Ids is part of the key source of overall segment threat insight
0
NEW Veeam Backup for Microsoft Office 365 1.5

With Office 365, it’s your data and your responsibility to protect it. NEW Veeam Backup for Microsoft Office 365 eliminates the risk of losing access to your Office 365 data.

 
SydNal2009Author Commented:
Do you know where I can find step-by-step, or something close to it, on how to configure a NIDS? Also, for the placement of the NIDS on the network, what  consideration should take into account if you are using a next generation firewall such SourceFire FirePower Appliance?
0
 
Rich RumbleSecurity SamuraiCommented:
The only consideration is basically how do you get the data you want to inspect inspected. Meaning, if you have 2-3 switches that handle traffic going in/out, how will you get that data to the IDS physically, and maybe how can you filter out unwanted noise/traffic.
With a cisco switch you use a span port, you tell it what vlans or what port's you want the span port to CC the IDS on, and that's that. If you have a few switches, then put in a few NIC's in the IDS, each NIC will get a swtich connected to the span port. Other switch makers call those ports monitor or mirror ports.

If you're placing the IDS on a certain floor of a building for example, but you need to get data from multiple floors (perhaps they don't all coalesce in one point), then you probably need a sensor on those other floors, and it can fwd it's DB to the central log.

The Snort and Suricata blogs have good information on them, it's not hard to put in place, it's the managment and tuning this always harder.
NGFW's or UTM's are not all that great. Despite many of their claims, the devil is in the details, because they ALL require a client installed on the host's so they know who is requesting what. You can't sniff http traffic and know the Jsmith is behind it, you need the client on the pc to tell the ngfw/utm that jsmith created that connection to xyz.com. UTM's and NGFW's gloss over these facts, there are many protocols out there that are essentially anonymous, your traffic looks just like my traffic, there is no differentiator unless we have explicitly put in a username and that can be seen, otherwise, just clicking around and downloading files etc.. there is no way for a NGFW or UTM to know who did that without a client on the PC's themselves telling the appliances.

https://redmine.openinfosecfoundation.org/projects/suricata/wiki
http://oreilly.com/catalog/snortids/chapter/ch06.pdf
-rich
0
 
btanExec ConsultantCommented:
Maybe useful for a start using sourcefire
https://community.sourcefire.com/questions/checklist-for-installs

The forum can search for more....
https://community.sourcefire.com/search?q=deploy+sourcefire

Just a note:
Sourcefire does not offer a state sharing HA configuration however there external tools made available from a couple of vendors that will provide redundancy.
0

Featured Post

Get free NFR key for Veeam Availability Suite 9.5

Veeam is happy to provide a free NFR license (1 year, 2 sockets) to all certified IT Pros. The license allows for the non-production use of Veeam Availability Suite v9.5 in your home lab, without any feature limitations. It works for both VMware and Hyper-V environments

  • 3
  • 2
Tackle projects and never again get stuck behind a technical roadblock.
Join Now