Oj Ruby Gem CVE-2026-54500
MEDIUMSeverity by source
AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:N/A:N
Network delivery of arbitrary JSON requires no authentication; impact is limited to low confidentiality (uncontrolled stack bytes); no integrity or availability effect.
Primary rating from Vendor (https://github.com/ohler55/oj).
CVSS VectorVendor: https://github.com/ohler55/oj
Lifecycle Timeline
2DescriptionCVE.org
Summary
Oj.load in :object mode reads uninitialized stack memory (and, for long keys, reads out of bounds) when parsing a JSON object whose key is 254 bytes or longer. The interned bytes can surface to the caller, disclosing process stack memory.
Details
In ext/oj/intern.c, form_attr() handles the long-key path by allocating a heap buffer b, populating it with the attribute name, and then freeing it - but it passed the uninitialized stack buffer buf (not b) to rb_intern3():
static VALUE form_attr(const char *str, size_t len) {
char buf[256];
if (sizeof(buf) - 2 <= len) { // long-key path (len >= 254)
char *b = OJ_R_ALLOC_N(char, len + 2);
// ... b is filled correctly ...
id = rb_intern3(buf, len + 1, oj_utf8_encoding); // BUG: reads `buf`
OJ_R_FREE(b);
return id;
}
// ...
}rb_intern3 therefore reads len + 1 bytes of uninitialized stack memory. When the key length is >= 256, it also reads out of bounds past the 256-byte buf (CWE-125). The resulting bytes are interned and can reach the caller via the produced Symbol or via the EncodingError message raised on invalid UTF-8, leaking process stack contents.
This is the same defect previously fixed in ext/oj/usual.c; intern.c held a duplicated copy of form_attr that was missed.
Proof of Concept
require 'oj'
key = "A" * 300
json = %Q[{"^o":"Object","#{key}":1}]
Oj.load(json, mode: :object)On affected versions this raises an EncodingError whose message contains ~1500 bytes of uninitialized stack memory (not the supplied "A"s). The leaked byte count varies between runs with the identical payload (e.g. 1491 vs 1516 bytes), confirming the content is uninitialized memory rather than fixed data.
Impact
Information disclosure of process stack memory to a caller that parses untrusted JSON with Oj.load(..., mode: :object). For keys >= 256 bytes it is also an out-of-bounds read (CWE-125).
Severity is bounded by several preconditions: it requires :object mode (which is already discouraged for untrusted input), the leaked bytes are uncontrolled (the attacker cannot choose what is disclosed), and the data only reaches an attacker if the application surfaces the resulting Symbol or EncodingError back to them. Scored CVSS 5.3 (Medium) on that basis.
Patches
Fixed in 3.17.3: form_attr() now passes b to rb_intern3 (a one-character change mirroring the earlier usual.c fix). Verified on the fixed build: the same payload returns cleanly with no leak across repeated runs.
Credit
Reported by Zac Wang (@7a6163).
AnalysisAI
Stack memory disclosure in the Oj RubyGem (all versions before 3.17.3) allows remote unauthenticated callers to recover uninitialized process stack contents by supplying a JSON object with keys of 254 bytes or longer to Oj.load in :object mode. The root cause is a copy-paste defect in ext/oj/intern.c where form_attr() allocates and fills a correct heap buffer b but then erroneously passes the uninitialized 256-byte stack buffer buf to rb_intern3(), causing leaked bytes to surface in the returned Symbol or an EncodingError message. A proof-of-concept is publicly available in the GHSA advisory; no confirmed active exploitation (CISA KEV) has been identified at time of analysis.
Technical ContextAI
Oj ('Optimized JSON') is a high-performance C-extension JSON parser for Ruby distributed as pkg:rubygems/oj. The vulnerable code resides in ext/oj/intern.c within form_attr(), which interns attribute names (as Ruby Symbols) during :object-mode parsing. For keys of 253 bytes or fewer, the function uses a local 256-byte stack buffer buf safely. When a key reaches 254 bytes (triggering the sizeof(buf) - 2 <= len branch), the function correctly heap-allocates a buffer b, populates it with the attribute name, but then calls rb_intern3(buf, len + 1, oj_utf8_encoding) - reading from the uninitialized stack buffer rather than b. For keys of 256 bytes or more, the read also extends past the end of buf, constituting a classic out-of-bounds read (CWE-125). This identical defect had been previously corrected in ext/oj/usual.c; intern.c contained a duplicated copy of form_attr that was missed in the earlier fix cycle. The interned bytes can surface to callers via the Symbol value or, when the stack bytes fail UTF-8 validation, via the text of the EncodingError exception.
RemediationAI
Upgrade Oj to version 3.17.3 or later, which corrects form_attr() in intern.c to pass the correctly populated heap buffer b to rb_intern3() - a one-character fix mirroring the earlier repair in usual.c. The fix is confirmed by the maintainer and documented at https://github.com/ohler55/oj/security/advisories/GHSA-fm7p-mprw-wjm9. If an immediate gem upgrade is not feasible, the most targeted compensating control is to replace mode: :object with :strict or :compat in all Oj.load calls that process untrusted input; note this changes deserialization behavior for callers relying on Ruby-object reconstruction. A secondary control is input validation that rejects JSON payloads containing keys of 254 bytes or more before they reach Oj.load, though this is fragile if not applied consistently at every entry point. Do not rely on error suppression alone as a mitigation, as the uninitialized read still occurs even if the EncodingError is rescued.
Same weakness CWE-125 – Out-of-bounds Read
View allSame technique Buffer Overflow
View allVendor StatusVendor
Share
External POC / Exploit Code
Leaving vuln.today