CVE-2026-59393
Patch Available.
This Vulnerability has been fixed in the Never-Ending Support (NES) version offered by HeroDevs.
Overview
Spring Framework is the core application framework of the Spring portfolio for Java, providing dependency injection, data access and transaction support, and the Spring MVC and Spring WebFlux web frameworks.
A denial of service vulnerability (CVE-2026-59393) has been identified in the Spring WebFlux XML decoding support, which allows attackers to make an application buffer XML input beyond the configured maxInMemorySize limit. A specially crafted XML payload can cause memory consumption to grow beyond that limit when the application decodes XML request or response bodies with the Aalto XML parser, potentially leading to denial of service.
Per OWASP: The Denial of Service (DoS) attack is focused on making a resource (site, application, server) unavailable for the purpose it was designed.
This issue affects multiple versions of Spring Framework.
Details
Module Info
- Product: Spring Framework
- Affected packages:
org.springframework:spring-webflux - Affected versions: <=5.2.26.RELEASE, >=5.3.0 <=5.3.50, >=6.0.0 <=6.0.32, >=6.1.0 <=6.1.29, >=6.2.0 <=6.2.20, >=7.0.0 <=7.0.10
- GitHub repository: https://github.com/spring-projects/spring-framework
- Published packages: https://central.sonatype.com/artifact/org.springframework/spring-webflux
- Package manager: Maven
- Fixed in:
- OSS Spring Framework 7.0.11
- NES for Spring Framework
Vulnerability Info
This Medium-severity vulnerability is found in the org.springframework:spring-webflux package in multiple versions of Spring Framework. WebFlux applications decode XML bodies through Jaxb2XmlDecoder, which uses XmlEventDecoder to turn incoming DataBuffer chunks into XML events. The decoder enforces maxInMemorySize to cap how much input may be held while a single XML element is being parsed. When the Aalto async XML parser is on the classpath, each buffer is fed to a non-blocking stream reader and released immediately, and the number of received bytes is counted by a ReceivedByteTracker:
public List<? extends XMLEvent> apply(DataBuffer dataBuffer) {
try {
if (this.byteTracker != null) {
this.byteTracker.incrementByteCount(dataBuffer);
}
AsyncByteBufferFeeder inputFeeder = this.streamReader.getInputFeeder();
try (DataBuffer.ByteBufferIterator iterator = dataBuffer.readableByteBuffers()) {
while (iterator.hasNext()) {
inputFeeder.feedInput(iterator.next());
}
}
List<XMLEvent> events = new ArrayList<>();
while (true) {
if (this.streamReader.next() == AsyncXMLStreamReader.EVENT_INCOMPLETE) {
// no more events with what currently has been fed to the reader
break;
}
else {
XMLEvent event = this.eventAllocator.allocate(this.streamReader);
events.add(event);
if (event.isEndDocument()) {
break;
}
}
}
return events;
}
...
}The tracker only counts bytes; the limit is checked later, by the caller that aggregates the emitted XML events. When a buffer ends inside a token that the parser cannot complete yet, such as an unterminated attribute value, comment, CDATA section or processing instruction, streamReader.next() returns EVENT_INCOMPLETE and the method returns an empty list. The parser keeps the partial token in its own internal buffers, but because no events are emitted, nothing downstream ever checks whether the tracked byte count has passed maxInMemorySize. An attacker who can send an XML body to an endpoint that decodes XML with WebFlux, or a malicious server answering a WebClient request, can open such a token and keep streaming data into it, so the parser accumulates input far beyond the configured limit and memory consumption grows until the application is degraded or runs out of memory.
Mitigation
Only recent versions of Spring Framework receive community support. Older lines are End-of-Life and will not receive public updates to address this issue.
Users of the affected components should apply one of the following mitigations:
- Upgrade to a currently supported version of Spring Framework.
- Leverage a commercial support partner like HeroDevs for post-EOL security support.
Credits
- h1ei1 (finder)
- YongHo Shin from eGovFrame Center (finder)