Take the 2-minute tour ×
Information Security Stack Exchange is a question and answer site for Information security professionals. It's 100% free, no registration required.

We have implemented/are implementing a network vulnerability scanning process, and we have chosen to use Qualysguard.

Qualys supply a scanner appliance for the internal network scanning, which obviously connects to the network. Our internal network is segmented into multiple subnets, each firewalled from the others. We have a network segment which is used for IT management (such as Snort and Splunk). It would seem to make sense to have the scanner appliance in this network segment. The obvious issue is that the scanner appliance needs full visibility into all the devices it will scan on the network, and I'm loathe to open up firewall ports if it can be helped. I know that someone will say that there should be a scanner appliance in each subnet, but that's not practical for us.

I can think of several possible network locations for the scanner appliance:

  • In the management network
  • At the most-protected segment of the network (scanning out to less-protected devices)
  • At the lest-protected segment of the network (scanning in to more protected devices)
  • In an entirely separate subnet

So my question is that given that the appliance needs full (all ports open) access to all devices, where would you put the scanner appliance on your network and why?

share|improve this question
1  
Isn't this a question for Qualys? –  schroeder Dec 21 '11 at 18:51
    
@schroeder, Qualys are good at telling you how to use their product, but don't give much guidance to more generalised vulnerability management, which is what I consider this question to be about. –  hmallett Feb 4 '12 at 22:30

2 Answers 2

up vote 2 down vote accepted

I have not used QualysGuard however using a vendor agnostic approach I would consider the following.

Indirect connection through firewall:

This can be a good option if you are unable to connect to the subnet for some reason (not enough connections, physically not accessible, etc...).

Some types of active scans depend on logical network location and may not work though a firewall / IPS (such a broadcasts and some poisoning) depending on your configuration. This is normally not a problem because the firewall / IPS provides that layer of protection, but an assessment may not be complete.

Increased load / processing through firewall and network equipment. Vulnerability scanning traffic can stress network equipment and may flood links. If this is a concern then locating the scanning interface logically close to the target is an advantage.

Direct connection to subnet:

Generally the best performance way to install a scanner, but can be a problem if you are unable to connect to a switch port with access to the network or VLAN you want to scan.

Directly connecting a scanner where it has direct connectivity to it’s target may allow it to detect vulnerabilities that are masked behind switch, firewall or IPS devices.

Scanners bridges secure zones. A scanner can have vulnerabilities too, less so these days but every scanner interface is potentially a bridge to multiple networks. This obviously depends on your implementation and product, some are better than others. Scanners would not normally route traffic but may if a vulnerability existed in the scanner or the scanner OS.

Conclusion:

The other options you have listed are all valid and could be used with an indirect connection through a firewall. Ideally you need to scan the platforms or applications from the side they will be servicing requests on. That may not be possible on a management network, depending on your configuration.

Where would you put the scanner appliance on your network and why?

I would connect the scanner with directly connected ports on your most secure subnets (Internal network) which would likely have more hosts to scan. And then put some other interfaces in a separate subnet with access through firewalls to your least secure subnets.

This limits bridging and allows you to scans most/all of your hosts with minimal performance hit.

Hope this helps.

share|improve this answer

I am not sure that you will be able to get away with not adding rules to your firewalls to allow the traffic from the scanner. You will not need to open them up globally, just for the scanner's IP.

If you are using stateful inspection, the scanners should be on the untrusted side of the firewall anyway so that the state tables of the firewalls do not fill up and cause traffic problems.

The QualysGuard scanner supports VLAN tagging, so you might be able to work around the firewalls, depending on your setup.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.