Skip to main content
Version: 10.0

AT22: Non-Spoofed UDP Flood

Definition​

During this attack, a victim server receives non-spoofed UDP packets at a very high packet rate and is overwhelmed by the large amount of incoming UDP packets. The attack consumes network resources and available bandwidth, exhausting the network until it shuts down. In Non-Spoofed UDP Flood packets, the source IP is the actual public IP of the attacker bot (or frequently, a public service being abused for reflection purposes), and the source IP range is equal to the number of bots or reflection targets used in the attack. This type of attack is harder to identify because it resembles good traffic.   

Attack Matrix Dimensions​

Nature of IPHandshakeSource IP RangePacket RatePacket SizePacket ContentFragmentingSession RateSession DurationVERB Rate
Non-Spoofed--SmallVery HighVector DependentValid--------

Defense Fundamentals​

In situations where an attacker has compromised a large server at a hosting center, they can launch a very large non-spoofed UDP flood attack with various kinds of payloads.   

RioRey's UDP algorithm watches for a large amount of UDP packets generated by a single source IP. The adjustment in UDP flood mitigation sets the number of ppm (packets per minute) that will be examined by a single source/destination IP UDP stream.   

Payload entropy algorithms are applied, and in cases of no entropy or excessive entropy, the source/destination IP UDP traffic will be rate limited to same rate specific.

Certain well known protocols known to be abused for reflection (such as SSDP, NTP and QOTD) also have special controls here.

Configuration Settings​

Click Edit to change default setting(s). When done, remember to Save before moving on to next mitigation section.  

FeatureDefault SettingComments
Filter ModeAuto (Auto/Monitor)Auto: Algorithms are executed and packets are filtered.
Monitor: Packets are viewed, counted and all algorithms are executed. No filtering of these packets will occur Monitor mode. DDoS traffic will be identified and represented in all attack-based charts and the Zone Traffic Chart will represent zero dropped traffic.
Off: No filtering or monitoring is performed.
Disable payload scoring (RIOS 5.3.0 or greater)UncheckedIf checked, all data payload tests are disabled. QOTD, SSDP, chargen, and NTP tests are still performed based on their settings.
Good payload score low threshold (RIOS 5.6.0 or greater)15
Good payload score high threshold (RIOS 5.6.0 or greater)95
Enable duplicate payload test (RIOS 5.6.0 orz greater)Checked
Allow QOTD traffic (RIOS 5.2.9 or greater)DisabledIf enabled, allows Quote of the Day (QOTD) traffic over UDP Port 17.
Allow SSDP traffic (RIOS 5.3.0 or greater)EnabledIf enabled, allows Simple Service Discovery Protocol (SSDP) traffic over UDP Port 1900.
Allow chargen traffic (RIOS 5.2.9 or greater)DisabledIf enabled, allows Character Generation Protocol (CHARGEN) traffic over UDP Port 19.
Validate NTP packets (RIOS 5.2.9 or greater)EnabledIf enabled, validates the NTP packet header and restricts NTP packets to the NTP versions allowed (Default versions are 2, 3 and 4) and NTP private mode request codes allowed (if it is an NTP private mode packet).
Block UDP port pairs (RIOS 7.0.0 or greater)NoneHitting Add creates a new row of two text boxes, symbolizing the Source Port to Destination port UDP traffic to drop. Using the "*" character symbolizes a wild card. Wild cards can be used in either the Source or Destination for a given entry, but not both (dropping all UDP traffic can be enabled in AT 62). Any UDP traffic matching the block port pair specified will be dropped as subtype 16 (shown below). Entries can be removed by hitting Delete. A maximum of 10 "Block" and "Allow" entries can be set up in a single Zone.
Only allow specified UDP port pairs (RIOS 7.0.0 or greater)DisabledEnabling this control enables entry of "Allow" Source to Destination port pairs. Unlike the "Block" specifications, only the traffic matching an "Allow" entry will be permitted in the zone. Traffic dropped due to not matching an "Allow" entry will be recorded as subtype 17 (shown below).
Allow UDP port pairs (RIOS 7.0.0 or greater)NoneHitting Add creates a new row of two text boxes, symbolizing the Source Port to Destination port UDP traffic that should be allowed in the zone. Using the "*" character symbolizes a wild card. Wild cards can be used in either the Source or Destination for a given entry, but not both. Any UDP traffic NOT matching an allow port pair specified will be dropped as subtype 17 (shown below). Entries can be removed by hitting Delete. A maximum of 10 "Block" and "Allow" entries can be set up in a single Zone.
Show NTP advanced controls (RIOS 5.2.9 or greater)DisabledEnabling this control creates a drop down menu showing you more options for NTP filtering that are used by Validate NTP Packets:
NTP versions allowed: Highlighted numbers are the versions of NTP traffic that are allowed.
NTP private mode request codes allowed: Highlighted numbers are the request codes that are permitted when it is an NTP private mode packet.

Subtypes​

Attack packets are classified under subtypes for better reporting. The subtype and reason the packet is labeled as invalid under its attack type can be found in the following table.

Attack TypeSubtypeDetails
AT221No payload data flood
AT222Payload data suspect (sequence)
AT223Payload data suspect (repeating)
AT224Payload data suspect (random)
AT225Stream data suspect (random, low)
AT226Payload data suspect (random, low)
AT227Stream data suspect (random, high)
AT228Payload data suspect (random, high)
AT229Duplicate payload flood
AT2210Duplicate stream payload flood
AT2211Attack stream packet
AT2216Source or destination port closed
AT2217Source or destination port not allowed
AT2221Invalid NTP Packet
AT2222NTP version or private mode request/response code not allowed