CVE-2026-112270
HIGH
7,1
Source: [email protected]
Attack Vector: adjacent
Attack Complexity: low
Privileges Required: none
User Interaction: none
Confidentiality: N/A
Integrity: N/A
Availability: N/A
MEDIUM
6,5
Source: [email protected]
Attack Vector: adjacent_network
Attack Complexity: low
Privileges Required: none
User Interaction: none
Scope: unchanged
Confidentiality: none
Integrity: none
Availability: high
Description
AI Translation Available
SGLang through 0.5.22 with the Mooncake disaggregation backend contains an uncontrolled memory allocation vulnerability in handle_staging_rsp() that lets unauthenticated network-adjacent attackers exhaust Prefill memory. Attackers can obtain the rank address from the /route endpoint and send STAGING_RSP messages with huge chunk_idx values under fresh room IDs, causing the scheduler to be OOM-killed.
EPSS (Exploit Prediction Scoring System)
Trend Analysis
EPSS (Exploit Prediction Scoring System)
Prevede la probabilità di sfruttamento basata su intelligence sulle minacce e sulle caratteristiche della vulnerabilità.
EPSS Score
0,0025
Percentile
0,2th
Updated
Single Data Point
Only one EPSS measurement is available for this CVE. Trend analysis requires multiple data points over time.
770
Allocation of Resources Without Limits or Throttling
IncompleteCommon Consequences
Security Scopes Affected:
Availability
Potential Impacts:
Dos: Resource Consumption (Cpu)
Dos: Resource Consumption (Memory)
Dos: Resource Consumption (Other)
Applicable Platforms
All platforms may be affected
https://github.com/sgl-project/sglang/blob/v0.5.22/python/sglang/srt/disaggrega…
https://github.com/sgl-project/sglang/blob/v0.5.22/python/sglang/srt/disaggrega…
https://github.com/sgl-project/sglang/issues/45014
https://www.vulncheck.com/advisories/sglang-through-0.5.22-memory-exhaustion-vi…