First, I must admit that I cannot write issues in English very well. The following content was generated with the help of AI, and I hope it helps you:
While fuzzing milo-parser, I found that certain malformed HTTP inputs can cause the parser to allocate too much memory and crash with OOM (out-of-memory).
All the oom-* artifacts I found are the same bug, just different inputs triggering the same infinite / large allocation logic.
These are the affected artifacts:
oom-3586b4fb729b567c760bd801e767fd94be98b804
oom-80d8972eb629edb24a95b8f4603ef7f0b4607976
oom-da39a3ee5e6b4b0d3255bfef95601890afd80709
This is a DoS vulnerability because an attacker can send a small, malicious HTTP message to make the parser consume all memory and crash.
The bug is in the HTTP parsing logic when handling certain invalid headers / values.
The parser tries to allocate an unbounded amount of memory without limits.
Thanks!
First, I must admit that I cannot write issues in English very well. The following content was generated with the help of AI, and I hope it helps you:
While fuzzing milo-parser, I found that certain malformed HTTP inputs can cause the parser to allocate too much memory and crash with OOM (out-of-memory).
All the oom-* artifacts I found are the same bug, just different inputs triggering the same infinite / large allocation logic.
These are the affected artifacts:
oom-3586b4fb729b567c760bd801e767fd94be98b804
oom-80d8972eb629edb24a95b8f4603ef7f0b4607976
oom-da39a3ee5e6b4b0d3255bfef95601890afd80709
This is a DoS vulnerability because an attacker can send a small, malicious HTTP message to make the parser consume all memory and crash.
The bug is in the HTTP parsing logic when handling certain invalid headers / values.
The parser tries to allocate an unbounded amount of memory without limits.
Thanks!