Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Application Security

6/26/2019
04:55 PM
Connect Directly
Twitter
LinkedIn
Google+
RSS
E-Mail
50%
50%

Developers and Security Teams Under Pressure to Collaborate

The challenges and benefits to getting two traditionally adversarial groups on the same page.

AWS re:Inforce – BOSTON – The path to secure development involves closer collaboration between the security and developer teams, a duo with a traditionally rocky relationship.

Application security, DevOps, and DevSecOps were all terms frequently heard this week at Amazon Web Services' re:Inforce, its first-ever security conference. AWS has been very developer focused, pointed out Chris Eng, chief research officer at Veracode. It's positive to see a focus on security's role in development, which he said has been a growing issue for four to five years.

Looking back 10- to 15 years, there has been a clear way security and development worked: engineers built code and handed it over to security when it was ready. By the time security came back with fixes, developers would be in a time crunch; rarely was there time to address them all. "There was a very rigorous, very structured set of handoffs," said Brian Riley, Liberty Mutual's senior director of global cyber risk management. "We're in a different world now."

Indeed, the transition to cloud broke down those choke points. DevOps evolved along with the cloud, driving the speed of new software releases and requiring security to review applications more frequently. This caused "a huge adjustment" for those who did application security a long time ago, Eng explained, because it shifted responsibility for security teams who assessed code.

"There's a tradeoff between depth and speed," he continued. "If I have a shorter amount of time to review something, there's a greater chance I'm going to miss something."

Developers must understand the types of things security will be looking for, he continued. Security, which has a reputation for holding progress back with constant fixes, has to meet the developer teams where they are and try not to disrupt what they're already doing. Security practitioners are traditionally uncomfortable with accepting risk. As the process of software development continues to accelerate, they will have to learn how to let some things go.

It was the move to cloud that prompted Riley, a former developer to collaborate with the dev team. "It challenged me, as a longtime security professional, to realize I had to get a lot closer to development," he explained. "I needed to be where the developers were." It wasn't – and still isn't – where security teams operate. Riley cited "drawn-out battles" between security and dev teams: security often says "that's not controlled;" devs respond with "this could be better."

"Historically, it's adversarial," Eng said of the longtime relationship between developers and security practitioners. "It's had to move from adversarial to more cooperative."

Security Champions: Bridging the Gap

Eng pointed to a growing pattern of "security champions," or developers with an aptitude in security who become an extension of the infosec team. These individuals are trained to conduct code reviews themselves as opposed to sending it off to security. The idea is to shift responsibility and lessen the workload for security teams, which are also focused on tasks outside secure development and often don't have the number of employees they need.

Of course, the appointment of security champions doesn't always sit well with security teams, he added. They need to hand off responsibility for code reviews to someone else; however, if something goes wrong, they're still to blame. Many are afraid of shifting this responsibility.

"There's a need to be more comfortable with losing a little bit of control," Eng said. If a dev team can handle 80% of security work, he added, it's helpful to the development process.

Overall, it also helps when developers have a security background, as it improves understanding between the two teams and, consequently, their working relationship. It's often not required for developers to have a security background, but it is a plus if they're interested in the space. Veracode does quarterly boot camps and exercises to train developers in cybersecurity.

This evolving collaboration signifies growing the decentralization of security, Eng said. Over time, he predicts, the back-and-forth between security and development will be erased.

Related Content:

 

Black Hat USA returns to Las Vegas with hands-on technical Trainings, cutting-edge Briefings, Arsenal open-source tool demonstrations, top-tier security solutions and service providers in the Business Hall. Click for information on the conference and to register.

Kelly Sheridan is the Staff Editor at Dark Reading, where she focuses on cybersecurity news and analysis. She is a business technology journalist who previously reported for InformationWeek, where she covered Microsoft, and Insurance & Technology, where she covered financial ... View Full Bio

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
Page 1 / 2   >   >>
tdsan
50%
50%
tdsan,
User Rank: Ninja
7/15/2019 | 3:58:36 PM
Re: spot on especially for container security


Out of the things mentioned in this article, they did not talk specifically about how it could affect IT Security/CyberSecurity. However, we could infuse the topics covered with CyberSecurity to Create Definitions -> Perform Data Extraction -> Data Classification -> Predict Historical Significance -> Pattern Recognition -> Regression Testing -> Predective Analytics; these areas of study can help to create an algorithm or policy to impede possible attacks. The items listed (applications to be used for ML) can be used to mitigate potential threats. The TIM (Threat Intelligence Machine) goes through a specific process that can be applied in numerous areas of IT, review the flow chart listed below:



Todd
pawan.shankar
50%
50%
pawan.shankar,
User Rank: Author
7/15/2019 | 2:54:05 PM
Re: spot on especially for container security
good call on the ML, I just wrote about why ML is a good fit for container security, https://www.darkreading.com/cloud/is-machine-learning-the-future-of-cloud-native-security/a/d-id/1335206
tdsan
50%
50%
tdsan,
User Rank: Ninja
7/15/2019 | 1:05:18 PM
Re: spot on especially for container security
Yes, agreed, we need to improve the application imaging process where the app is locked down using a Pentesting process and an app policy needs to be activated when referring to the internal functions, characteristics and outcomes should be studied so a policy change can be put in place to mitigate organizational problems:
  • "yum install selinux; systemctl enable selinux; systemctl state selinux" # Install, enable and verify state
  • "grep -i avc /var/log/audit/audit.log | audit2allow --module local > local.te" # Create module for something specific, review the policy, I am looking the web traffic since this is a web server
  • "yum install policycoreutils-devel" # Gives users the ability to configure policies from the desktop
  • "find / -name "local.te" print0 | xargs -0 -t semodule -i" # looks for local.te and installs it as one of the modules
  • "setsebool -P httpd_read_user_content" # Sets the policy permanently to read pages that have been served up by nginx or apache

There are other application programming aspects that we can review but this gives the user an idea of how a developer and security consultant can work together to manipulate the characteristics of files, filesystem, domains, policies, and executables.

Todd
tdsan
50%
50%
tdsan,
User Rank: Ninja
7/11/2019 | 12:15:12 PM
Re: spot on especially for container security
I agree with the points you made, I do think containers are the way, in addition:
  • SELinux - we need to work with the security teams and DevOps team to secure the application, filesystem, domain, executable/bin and application characteristics
  • "setenforce 1; reboot; getenforce; yum install nginx; systemctl start nginx; semanage fcontext -a -R 'httpd_sys_context_t' '/usr/share/nginx/html(/.*)?'; setsebool -P http_read_user_content 1; semanage port -a -t 'http_port_t -p tcp 80,443 (this will limit what the external user has access to)"
  • "iptables -I INPUT 1 -p tcp -m multiport --dport 80,443 -m conntrack --ctstate NEW -j ACCEPT"
  • "for i in `rpm -qa | grep apache`; rpm -e $i ; done #(remove apache from the system (nginx is installed now)"
  • "Docker configuration - yum install docker; systemctl start docker; docker login <enter user from docker hub>; docker pull nginx; docker run -p 8082:80 --name nginx2 -v $PWD:/usr/share/nginx/html/docs --restart unless-stopped nginx # this creates a docker container on your linux machine running nginx latest image or richarvey\nginx-php-fpm # works well using secure https" 
  • Firewall - we need to start utilizing NGFW where the FW is able to think and look at the application's heuristics and understand the application. If there is an outside variant (i.e. Ransomware) and it activates encryption tools, the NGFW (onboard or external) needs to put that process in a container or isolate that process where its effects are limited, what can DevOps do to work with the Security teams to create a pre and post scenario when it finds this type of attack (collaborate to resolve this issue)
  • ML - Machine Learning is missing from this equation but with the DevOps teams help by using Tensorflow techniques, we can train the ML aspect of the application to look for anomalies and variants (especially those that access ssh/keys over the internet after numerous attempts)

Todd
pawan.shankar
50%
50%
pawan.shankar,
User Rank: Author
7/8/2019 | 11:11:54 AM
spot on especially for container security
this seems to be very applicable in the container security space, where devops and security are traditionally in different silos. Teams need to work closer together and the pressure to collaborate is increasing in order to not sacrifice agility for security and vice versa.
tdsan
50%
50%
tdsan,
User Rank: Ninja
6/29/2019 | 1:51:43 PM
Re: Control
Interesting, I do think it depends on the type of work they are performing. If someone from the security team can review the code to determine if the code meets 'CMMI' coding best practices at the beginning of the Agile scrum development process, then yes I agree but that is dependent on the level of compentency the security team has when it comes to "SecDev" or the "Security Development process.
However, in most cases, security groups don't have coders on the team with a strong background in security, so I do think this aspect of the industry is evolving, by combining these business areas, in the long-run it will benefit the industry.

Todd
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
6/28/2019 | 6:35:23 PM
Control
If a dev team can handle 80% of security work, he added, it's helpful to the development process. This is the case in most of the situations but it can be managed.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
6/28/2019 | 6:31:57 PM
Trade-off
There's a tradeoff between depth and speed This is really true. Balancing act.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
6/28/2019 | 6:30:27 PM
Cloud
DevOps evolved along with the cloud, driving the speed of new software releases and requiring security to review applications more frequently. Agree. Cloud made it more possible to engage security team early enough, as they enforce certain rules from the beginning.
Dr.T
50%
50%
Dr.T,
User Rank: Ninja
6/28/2019 | 6:28:32 PM
Re: SecDev - wave of the future
We will be incorporating "SecDev" or "Security Development" teams into our environment as well and thank you for this valuable insight. Agree this will be important in short and long terms.
Page 1 / 2   >   >>
US Turning Up the Heat on North Korea's Cyber Threat Operations
Jai Vijayan, Contributing Writer,  9/16/2019
Preventing PTSD and Burnout for Cybersecurity Professionals
Craig Hinkley, CEO, WhiteHat Security,  9/16/2019
NetCAT Vulnerability Is Out of the Bag
Dark Reading Staff 9/12/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Current Issue
7 Threats & Disruptive Forces Changing the Face of Cybersecurity
This Dark Reading Tech Digest gives an in-depth look at the biggest emerging threats and disruptive forces that are changing the face of cybersecurity today.
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-16413
PUBLISHED: 2019-09-19
An issue was discovered in the Linux kernel before 5.0.4. The 9p filesystem did not protect i_size_write() properly, which causes an i_size_read() infinite loop and denial of service on SMP systems.
CVE-2019-3738
PUBLISHED: 2019-09-18
RSA BSAFE Crypto-J versions prior to 6.2.5 are vulnerable to an Improper Verification of Cryptographic Signature vulnerability. A malicious remote attacker could potentially exploit this vulnerability to coerce two parties into computing the same predictable shared key.
CVE-2019-3739
PUBLISHED: 2019-09-18
RSA BSAFE Crypto-J versions prior to 6.2.5 are vulnerable to Information Exposure Through Timing Discrepancy vulnerabilities during ECDSA key generation. A malicious remote attacker could potentially exploit those vulnerabilities to recover ECDSA keys.
CVE-2019-3740
PUBLISHED: 2019-09-18
RSA BSAFE Crypto-J versions prior to 6.2.5 are vulnerable to an Information Exposure Through Timing Discrepancy vulnerabilities during DSA key generation. A malicious remote attacker could potentially exploit those vulnerabilities to recover DSA keys.
CVE-2019-3756
PUBLISHED: 2019-09-18
RSA Archer, versions prior to 6.6 P3 (6.6.0.3), contain an information disclosure vulnerability. Information relating to the backend database gets disclosed to low-privileged RSA Archer users' UI under certain error conditions.