Surge in Hacktivist DDoS: Keymous+ and DieNet Activity Analysis
Has anyone else noticed the spike in noise on their edge sensors following the recent geopolitical developments? Radware released a report this morning detailing a massive wave of retaliatory DDoS attacks after the 'Epic Fury' and 'Roaring Lion' campaigns.
According to the data, 149 attacks hit 110 organizations across 16 countries. What caught my eye is that this isn't just a disorganized swarm; two groups—Keymous+ and DieNet—are responsible for nearly 70% of the malicious traffic between Feb 28 and March 2.
While volumetric attacks are nothing new, the coordination here is notable. We've started seeing patterns indicating HTTP floods trying to exhaust server resources rather than just bandwidth pipe saturation.
I've pushed some updated rate-limiting rules to our NGINX reverse proxies to mitigate the immediate risk. If you're seeing similar traffic, you might want to enforce stricter burst limits in your config:
nginx limit_req_zone $binary_remote_addr zone=one:10m rate=10r/s;
server { location / { limit_req zone=one burst=20 nodelay; } }
For those analyzing logs, look for high-frequency requests from single IPs that might be slipping under standard WAF thresholds. You can use this quick awk snippet to isolate top talkers:
awk '{print $1}' /var/log/nginx/access.log | sort | uniq -c | sort -nr | head -n 10
Are you guys seeing specific payload signatures, or is this mostly brute-force volume in your environment?
We've seen a mix of both. While the volumetric attacks are loud, we noticed DieNet specifically targeting API endpoints with malformed JSON payloads to trigger 500 errors. It’s an attempt to saturate the application thread pool rather than the network card.
If you're using AWS WAF, I recommend deploying a rate-based rule specifically on your API gateway paths immediately.
Good catch on the User-Agent strings. We actually pivoted to fingerprinting based on TLS fingerprinting (JA3) because these groups often rotate user agents but keep the same SSL stacks.
I wrote a quick Python script to pull JA3 hashes from Zeek logs and cross-reference them against known botnet signatures. It's been significantly more effective than just looking at IP reputation for this campaign.
Verified Access Required
To maintain the integrity of our intelligence feeds, only verified partners and security professionals can post replies.
Request Access