Keep IoT devices and data safe from threats
- By Christopher Camejo
- Jun 01, 2017
The risk posed by hackers to the Internet of Things
(IoT) is a hot topic and there have already been some
serious real-world attacks. Any organization deploying
network-connected devices would be well advised to
take a deeper look at the risks posed and take steps to
When we talk about any risk we must consider the frequency with
which incidents may occur and the impact that an incident would
have on the organization. Some events may be fairly common but
carry a low impact while others are rarer but carry tremendous consequences.
Determining what we realistically should be worried about
will help decide the measures that are appropriate to defend an IoT
installation against attacks.
We must consider what the IoT device itself is being used for.
Many IoT devices are used for physical security, such as cameras and
door access control, and if they were compromised could be expected
to lead to a compromise of physical security, which may be severe
depending on the type of facility. Moving to the higher impact end of
the spectrum, most hospitals now use medical devices that are connected
to the network and a breach of those devices could potentially
be a matter of life and death for the patients relying on them. On
a broader scale, utilities are using more IoT devices for monitoring
and management of infrastructure. A successful attack could impact
many thousands or even millions of people.
The direct risk posed by the devices themselves may be serious but
it is only one aspect of their risk. As modern IoT devices contain as
much computing power as personal computers from only a few years
ago and are often based on the same underlying architectures and
software, they can also be leveraged to attack other network resources.
A poorly protected IoT device could be used to attack a sensitive
server or database and then exfiltrate the resulting data.
Many organizations may believe they have nothing of value on
their network that an attacker would want (I would disagree, attackers
always want bank login credentials), but even in this case an attacker
can leverage compromised IoT devices to attack other organizations.
We saw this in 2016 with the Mirai botnet, which was primarily based
on IP cameras that had default or backdoor passwords; the resulting
attacks knocked many of the largest websites offline.
Protecting IoT devices requires efforts from both the vendors that
create and sell these devices and the users who install them. If a device
isn’t designed properly there will be no way to secure it regardless
of what steps the user takes, while even products with the best builtin
security will require some effort on the part of users during the
IoT devices need to be designed with security “baked in”. This means
using proper authentication and encryption for both data transmission
and administration. Other common traps need to be avoided as
well, like using the same default password for every device (users are
unlikely to change them) or, even worse, creating a backdoor account
with a hardcoded password (these are fairly easy for attackers to find
and extremely easy to exploit once found).
Vendors also have to consider the lifespan of their products. Many
consumer product companies have had a “sell it and forget it approach,”
which presents an enormous risk when software and network
connectivity is in play. A vulnerability in a 10-year-old product is potentially
more valuable to an attacker than a vulnerability in a just-released
product, as the installed base of the older product is likely to be much
larger. Vendors must take reports of security vulnerabilities seriously
and release patches, even for long-discontinued products.
The devices themselves will require some security configuration. Removing
default passwords and replacing them with strong passwords
is one of the most important steps, but other settings should be reviewed
as well. Depending on the device these could include encryption
settings, lists of IP addresses that are allowed to administer the
device, and authorizations for interactions with other devices.
The devices themselves will also require maintenance and monitoring
just like a PC. Patches should be applied quickly in order to
address security vulnerabilities and the network traffic generated by
the devices should be monitored to make sure that they have not been
co-opted for use by an attacker.
The devices themselves should also be protected, just like any other
computer. This means firewalling them off from the Internet. Search
engines, like the one available at shodan.io, make it easy for attackers
to search for vulnerable devices that are exposed.
Many IoT devices leverage cloud resources. This may be for remote
storage of data or to enable easy administration via a web frontend.
It’s important to remember that “cloud” resources are just another
set of computers in a datacenter with all of the same security concerns
of a regular computer, only in this case the user is relying on the
cloud provider to handle the security.
Users of cloud services should once again consider the risk
posed by the compromise or loss of any data that they are about to
place in the cloud and make sure that contracts
with cloud providers include provisions for an
appropriate level of security. If vendors aren’t
willing to commit to securing the data then
cloud services may not be the right fit.
This article originally appeared in the June 2017 issue of Security Today.