Web Application Attack Report … US is primary source of attacks

Imperva released a Web Application Attack Report that provides an overview of the principal cyber security events occurred during the last 12 months.

Imperva security firm has just released last annual Web Application Attack Report (WAAR) that provides an overview of the principal cyber security events occurred in the cyberspace during the last 12 months.

The Web Application Attack Report  illustrates that web applications are constantly under attack and that the majority of the offensives are originated from sources within the US.

The scenario is apocalyptic, it is improperly defined as a cyber war context by many security experts and the events and the damage caused by the attackers continue to be very concerned.

The report explores the dynamics within a riot cyberspace, the methods , the magnitude and the duration of the cyber attacks all over the world. The Web Application Attack Report provides a useful overview on the evolution of principal cyber threats and of the targets hit during the last year and reveals which countries perpetrated the most attacks.

Following the Key findings:

Retailers suffer twice as many SQL injection attacks as other industries

The following table shows that SQL injection attacks on retail applications were more intense, both in terms of number of attacks per incident and duration of an incident. Retail applications suffered 749 individual attack requests per attack campaign, a possible explanation is that these applications are characterized by a large surface of attack.

Many websites are constantly under attack

Imperva researchers observed that one observed website was under attack 176 out of 180 days, that correspond to 98% of the time.  A typical application suffered at least one cyber attack in 12 days, despite many applications are constantly under attack, the worst scenario observer was related to a company that was attacked attacked on average as many as 26 times per minute. The duration of the attacks is very variable, the typical attack incident lasted around five minutes meanwhile the worst-case incident was lasting more than 15 hours.

“In the above, it is evident that attacks are usually a burst event, with the medians number of attack incidents lasting between 5 to 10 minutes. As the maximal values show, there certainly are longer attacks lasting several hours, but most attacks are well below 15 minutes.”

The Web Application Attack Report indicates the US as the primary source of attacks

The US retains its rank as the number one source of web attacks, Imperva experts noted that the majority of offensives originated in the United States, Western European countries, China and Brazil.

“We believe that, with the current threat landscape, organizations can no longer afford to take an every-man-for-himself approach to security,” “This report demonstrates that the automation and scale of attacks leave a large footprint that can be better addressed by looking at data gathered from a large set of potential victims. Thus it is important to rely on one’s peers to acquire intelligence on malicious sources and apply this intelligence in real time.” said Amichai Shulman, CTO at Imperva.

Researchers analyzed the geographic dispersion of the attack initiating hosts, as determined by their IP addresses.

 

The Web Application Attack Report shows that there were no meaningful differences in countries of origin between the retail group and the rest of the analyzed application.

“For business logic attacks, email extraction is still widely dominated by African countries, such as Senegal, Nigeria, Ghana, and the Ivory Coast. The field of comment spamming shows an unusual proportion toward Eastern European countries, such as Russia, Ukraine, Latvia, and Poland. Besides this geographic dominance, more and more business logic attacks are originating from Asia and South America. Countries generating an increasing number of business logic attacks include Malaysia, Thailand, Pakistan, Mexico, Brazil, and Argentina.”

The report has in its last part a section dedicated to “Conclusions and Recommendations” for attack mitigation, principal issues are:

• Deploy security solutions that prevent automated attacks that are able to differentiate between automated bots and human clients. Security procedures and solutions should be as automated as possible to mitigate an attack volume that is too overwhelming for humans to monitor.
• Share information and threat intelligence about cyber attacks.
• Detect and block attacks that target known vulnerabilities.
• Acquire intelligence on malicious sources to blacklist principal attack sources.
• Estimate countermeasures on the worst case scenario, not on the average case.

Pierluigi Paganini

(Security Affairs – cybersecurity, Web Application Attack Report)