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.

Vulnerabilities / Threats

2/27/2019
05:52 PM
Connect Directly
Twitter
LinkedIn
RSS
E-Mail
50%
50%

More Than 22,000 Vulns Were Disclosed in 2018, 27% Without Fixes

As in previous years, input validation vulnerabilities accounted for a substantial proportion of total, Risk Based Security report shows.

The number of security vulnerabilities present in web applications and other software shows little signs of coming down anytime soon.

A new report from Risk Based Security (RBS) shows that a total of 22,022 security vulnerabilities were disclosed in 2018. The number is projected to increase substantially once vulnerabilities that were disclosed after the report was compiled are added to the total. Last year for instance RBS initially reported a total of 20,832 vulnerabilities. It later revised that figure to 22,230 after accounting for flaws in 2017 that were disclosed after the report was ready.

The main takeaway for enterprises is to remain vigilant, says Brian Martin, vice president of vulnerability intelligence at RBS. "Vulnerabilities are still being disclosed in high numbers, and every type of software is impacted," he notes. "Organizations, regardless of size, need to be aware of the vulnerabilities and constantly enhance their triage process."

As has been the case in recent years, web-related vulnerabilities accounted for nearly half (47.9%) of all reported security flaws in 2018. Some 27.5% were vulnerabilities tied to access authentication and 3.5% were classified as SCADA vulnerabilities—a doubling from 2017.

Input validation vulnerabilities—such as SQL injection errors, cross-site scripting, buffer overflow and command injection—once again accounted for a substantial majority of disclosed bugs in 2018. More than two-thirds of reported vulnerabilities (67.7%) last year were related to insufficient or improper validation of input suggesting that developers are still struggling to address an issue that has topped OWASPs list of top 10 vulnerabilities for a long time. Bug bounty programs, including those managed by vendors accounted for almost 8% of reported vulnerabilities in 2018, compared to just 5.8% the year before.

Of the total number of disclosed bugs last year, approximately 33% received a severity rating of seven above. Nearly one-third of them had public exploits available and slightly more than half were remotely exploitable. But, for the third year in a row, software vulnerabilities with a severity rating of between 9 and 10—the highest risk category—declined as a proportion to 13.6% of the overall total.

The reason for it could simply be that researchers are publishing a higher percentage of lower-risk vulnerabilities than before, Martin says. "If more researchers publish XSS, CSRF, or path disclosures—all considerably lower than 9.0—that could cause the percentage of lower-scored vulnerabilities to increase," even as the actual number of high-risk flaws increase.

It is also likely that some high-severity security vulnerabilities are also being deliberately kept quiet, though that number is likely insignificant, Martin says. "While governments do keep some 0-day vulnerabilities that are typically 9.3 or 10.0 scores, there has been no indication that they are sitting on hundreds of them at any given time. It could be the case, but we only have a few samples to go off," he says.

Significantly, RBS' report shows that organizations relying solely on the CVE / National Vulnerability Database (NVD) for their vulnerability information are missing a substantial number of bugs. In total, RBS had 6,780 more vulnerabilities in its database compared to the NVD. Of that number, nearly 46% had a severity rating of 7 or higher. "We have a very different mindset and philosophy when it comes to aggregating vulnerabilities," Martin explains. "We actually go looking for the vulnerabilities," rather than waiting for bugs to be reported, he notes. Others have noted the same issue. In 2017, research conducted by Recorded Future showed that more than 75% of vulnerabilities are disclosed online publicly before the NVD included them in its database.

Troublingly for organizations, RBS' data showed that nearly three-in-10 flaws reported in 2018 (27.1%) had no known or available fixes. That statistic highlights the reason why organizations need to have a defense-in-depth model, Martin notes. The goal should be to make vulnerable systems more difficult to access using approaches like access control lists, network segregation and technologies such as IDS and IPS, he said.

Related Content:

  

 

 

Join Dark Reading LIVE for two cybersecurity summits at Interop 2019. Learn from the industry's most knowledgeable IT security experts. Check out the Interop agenda here.

Jai Vijayan is a seasoned technology reporter with over 20 years of experience in IT trade journalism. He was most recently a Senior Editor at Computerworld, where he covered information security and data privacy issues for the publication. Over the course of his 20-year ... View Full Bio
 

Recommended Reading:

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
ChadF3
50%
50%
ChadF3,
User Rank: Strategist
3/4/2019 | 5:34:35 PM
Not about validation
Issues like SQL/script injection and XSS are not really a problem with input validation, but a lack of proper quoting/escaping of arbatrary data during its use. Yes, input validation can limit/mitigate some attacks, but this also tightly couples the input code and processing code (e.g. the input side assumes SQL queries are used at some point, and the query side code assumes input validation have done for all potential sources).

In general, providing an input with an apostrophe should be considered valid, even if the backend uses this value in an SQL query (as the real data being search may have such values, e.g. Name: O'Neill). If I include the text "<grin>" with-in some online forum post/comment, it shouldn't be rejected/removed for no-HTML input fields, but be treated as verbatuim text that is properly escaped in output HTML.

Input validation should normally only be done for specific business logic (e.g. usernames are limited to ASCII letters/numbers and must start with a letter), or general sanity checks (e.g. only ASCII; only valid UTF-8 encoded strings; doesn't contain invalid UNICODE composed sequences; no embedded NUL characters [when using length specified strings]).
Joe Stanganelli
50%
50%
Joe Stanganelli,
User Rank: Ninja
2/28/2019 | 6:31:14 PM
No fixes
Moreover, researchers are often incentivized against responsible disclosure simply because the companies won't even listen to them and give them the time of day -- unless it gets national media attention. (Examples: Panera, Apple FaceTime bug, Facebook, etc.)
Attackers Leave Stolen Credentials Searchable on Google
Kelly Sheridan, Staff Editor, Dark Reading,  1/21/2021
How to Better Secure Your Microsoft 365 Environment
Kelly Sheridan, Staff Editor, Dark Reading,  1/25/2021
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win an Amazon Gift Card! Click Here
Latest Comment: We need more votes, check the obituaries.
Current Issue
2020: The Year in Security
Download this Tech Digest for a look at the biggest security stories that - so far - have shaped a very strange and stressful year.
Flash Poll
Assessing Cybersecurity Risk in Today's Enterprises
Assessing Cybersecurity Risk in Today's Enterprises
COVID-19 has created a new IT paradigm in the enterprise -- and a new level of cybersecurity risk. This report offers a look at how enterprises are assessing and managing cyber-risk under the new normal.
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2021-3317
PUBLISHED: 2021-01-26
KLog Server through 2.4.1 allows authenticated command injection. async.php calls shell_exec() on the original value of the source parameter.
CVE-2013-2512
PUBLISHED: 2021-01-26
The ftpd gem 0.2.1 for Ruby allows remote attackers to execute arbitrary OS commands via shell metacharacters in a LIST or NLST command argument within FTP protocol traffic.
CVE-2021-3165
PUBLISHED: 2021-01-26
SmartAgent 3.1.0 allows a ViewOnly attacker to create a SuperUser account via the /#/CampaignManager/users URI.
CVE-2021-1070
PUBLISHED: 2021-01-26
NVIDIA Jetson AGX Xavier Series, Jetson Xavier NX, TX1, TX2, Nano and Nano 2GB, L4T versions prior to 32.5, contains a vulnerability in the apply_binaries.sh script used to install NVIDIA components into the root file system image, in which improper access control is applied, which may lead to an un...
CVE-2021-1071
PUBLISHED: 2021-01-26
NVIDIA Tegra kernel in Jetson AGX Xavier Series, Jetson Xavier NX, TX1, TX2, Nano and Nano 2GB, all L4T versions prior to r32.5, contains a vulnerability in the INA3221 driver in which improper access control may lead to unauthorized users gaining access to system power usage data, which may lead to...