Detect Direct Network Flood in Microsoft Sentinel
Adversaries may attempt to cause a denial of service (DoS) by directly sending a high-volume of network traffic to a target. Direct Network Floods use one or more systems to send high-volume network packets toward the targeted service or network. Any network protocol may be used — stateless protocols such as UDP and ICMP are common due to their low overhead, but TCP SYN floods are also prevalent. Botnets are frequently leveraged to amplify attack volume, with compromised endpoints acting as unwitting flood sources. Organizations may detect this technique either as a victim observing inbound traffic spikes, or by identifying compromised endpoints in their environment participating in an outbound DDoS campaign as botnet nodes.
MITRE ATT&CK
- Tactic
- Impact
- Technique
- T1498 Network Denial of Service
- Sub-technique
- T1498.001 Direct Network Flood
- Canonical reference
- https://attack.mitre.org/techniques/T1498/001/
KQL Detection Query
// T1498.001 — Direct Network Flood
// Detects known DDoS/flood tool execution on endpoints (botnet participants, insider threat, red team)
let FloodToolNames = dynamic([
"hping3", "hping", "nping", "trafgen", "t50",
"loic", "hoic", "mhddos", "ufonet", "goldeneye", "xerxes",
"udpflood", "synflood", "icmpflood", "pyflood", "rudy",
"packetsender", "ostinato", "netcat"
]);
let FloodArgPatterns = dynamic([
"--flood", "-i u0", "--rand-dest", "--rand-source",
"--syn --flood", "--icmp --flood", "--udp --flood",
"-c 999999", "--count 9999999", "--faster", "--turbo",
"--rate 100000"
]);
DeviceProcessEvents
| where Timestamp > ago(24h)
| where FileName has_any (FloodToolNames)
or ProcessCommandLine has_any (FloodToolNames)
or ProcessCommandLine has_any (FloodArgPatterns)
| extend IsKnownFloodTool = FileName has_any (FloodToolNames)
or ProcessCommandLine has_any (FloodToolNames)
| extend HasFloodArgs = ProcessCommandLine has_any (FloodArgPatterns)
| extend SuspicionLevel = case(
IsKnownFloodTool and HasFloodArgs, "Critical",
IsKnownFloodTool, "High",
HasFloodArgs, "Medium",
"Low")
| project Timestamp, DeviceName, AccountName, FileName, ProcessCommandLine,
InitiatingProcessFileName, InitiatingProcessCommandLine,
IsKnownFloodTool, HasFloodArgs, SuspicionLevel
| sort by Timestamp descDetects Direct Network Flood activity by identifying known DDoS and flood tool execution on endpoints using DeviceProcessEvents. Matches against a curated list of tool names (hping3, LOIC, MHDDoS, ufonet, goldeneye, etc.) in both the process image name and command line, as well as flood-specific argument patterns (--flood, --rand-dest, high packet count flags). Assigns a SuspicionLevel to help analysts prioritize: Critical when both a known tool name and flood arguments are present, High for known tool name only, Medium for flood arguments from an unknown binary.
Data Sources
Required Tables
False Positives & Tuning
- Load testing tools (Apache Bench, wrk, hey, vegeta, k6) run legitimately by QA engineers or DevOps teams against internal staging environments or authorized external endpoints
- Network performance benchmarking tools (iperf, iperf3, netperf, iperf) used by infrastructure teams to validate bandwidth capacity on new circuits or after changes
- Security scanners using aggressive timing profiles (nmap -T5, masscan) operated by internal vulnerability management or red team programs with authorized change tickets
- Authorized DDoS simulation exercises where security vendors run flood tools against hardened DMZ systems to validate WAF or DDoS protection efficacy
- Academic, research, or sandbox environments where flood tools are used for network research, coursework, or tool development without malicious intent
Other platforms for T1498.001
Testing Methodology
Validate this detection against 4 adversary techniques from Atomic Red Team. Each test below lists the behaviour to exercise and the telemetry you should expect to see. Executable commands and cleanup steps are available with Pro.
- Test 1UDP Flood with hping3 to Localhost
Expected signal: Linux auditd: syscall execve for hping3 with argv containing '--udp', '--flood', '-p 53', '127.0.0.1'. Sysmon for Linux (if deployed): Process execution event with Image=hping3 and full CommandLine. Network metrics: High-volume UDP packet rate on loopback interface visible in netstat -s and /proc/net/udp. The -c 10000 flag limits total packets to prevent resource exhaustion.
- Test 2PowerShell UDP Packet Burst to Localhost
Expected signal: Sysmon Event ID 1: Process Create for powershell.exe with CommandLine containing 'UdpClient', 'Send', 'Loopback'. Sysmon Event ID 3: High-frequency network connection events to 127.0.0.1:53 — this will generate thousands of Event ID 3 records in the Sysmon log. PowerShell ScriptBlock Log Event ID 4104 with full script content if ScriptBlock logging is enabled.
- Test 3ICMP Flood with ping -f to Localhost
Expected signal: Linux auditd: syscall execve for ping with argv '-f', '-c', '10000', '127.0.0.1'. Process execution in syslog or Sysmon for Linux. Output shows packet statistics: '10000 packets transmitted, 10000 received, 0% packet loss'. Network metrics show ICMP packet rate spike on loopback interface visible via /proc/net/snmp ICMP counters.
- Test 4LOIC Flood Tool Binary Staging and Execution Simulation
Expected signal: Sysmon Event ID 11: File Create event for %TEMP%\loic.exe — triggers the file-staging hunting query on flood tool binary name. Sysmon Event ID 1: Process Create for loic.exe with CommandLine '--method udp --target 127.0.0.1 --port 80 --threads 10'. Security Event ID 4688 (if command-line auditing enabled) with same process details. Sysmon Event ID 7: Image Load events showing DLLs loaded by the loic.exe process.
References (8)
- https://attack.mitre.org/techniques/T1498/001/
- https://www.cisco.com/c/en/us/td/docs/ios-xml/ios/netflow/configuration/15-mt/nf-15-mt-book/nf-detct-analy-thrts.pdf
- https://www.justice.gov/opa/pr/seven-iranians-working-islamic-revolutionary-guard-corps-affiliated-entities-charged
- https://github.com/redcanaryco/atomic-red-team/blob/master/atomics/T1498.001/T1498.001.md
- https://learn.microsoft.com/en-us/azure/ddos-protection/ddos-protection-overview
- https://learn.microsoft.com/en-us/defender-endpoint/advanced-hunting-devicenetworkevents-table
- https://learn.microsoft.com/en-us/defender-endpoint/advanced-hunting-deviceprocessevents-table
- https://learn.microsoft.com/en-us/defender-endpoint/advanced-hunting-devicefileevents-table
Unlock Pro Content
Get the full detection package for T1498.001 including response playbook, investigation guide, and atomic red team tests.