Impact

Unix-like operating systems may segfault due to dereferencing a dangling pointer in specific circumstances. This requires an environment variable to be set in a different thread than the affected functions. This may occur without the user's knowledge, notably in a third-party library.

Workarounds

No workarounds are known.

References

• time-rs/time#293

On Windows, configuring a named pipe server with pipe_mode will force ServerOptions::reject_remote_clients as false.

This drops any intended explicit configuration for the reject_remote_clients that may have been set as true previously.

The default setting of reject_remote_clients is normally true meaning the default is also overridden as false.

Workarounds

Ensure that pipe_mode is set first after initializing a ServerOptions. For example:

let mut opts = ServerOptions::new();
opts.pipe_mode(PipeMode::Message);
opts.reject_remote_clients(true);

A bug introduced in rustls 0.23.13 leads to a panic if the received TLS ClientHello is fragmented. Only servers that use rustls::server::Acceptor::accept() are affected.

Servers that use tokio-rustls's LazyConfigAcceptor API are affected.

Servers that use tokio-rustls's TlsAcceptor API are not affected.

Servers that use rustls-ffi's rustls_acceptor_accept API are affected.

Affected version of this crate did not properly parse unknown fields when parsing a user-supplied input.

This allows an attacker to cause a stack overflow when parsing the mssage on untrusted data.

Summary

The DNSSEC validation routines treat entire RRsets of DNSKEY records as trusted once they have established trust in only one of the DNSKEYs. As a result, if a zone includes a DNSKEY with a public key that matches a configured trust anchor, all keys in that zone will be trusted to authenticate other records in the zone. There is a second variant of this vulnerability involving DS records, where an authenticated DS record covering one DNSKEY leads to trust in signatures made by an unrelated DNSKEY in the same zone.

Details

verify_dnskey_rrset() will return Ok(true) if any record's public key matches a trust anchor. This results in verify_rrset() returning a Secure proof. This ultimately results in successfully verifying a response containing DNSKEY records. verify_default_rrset() looks up DNSKEY records by calling handle.lookup(), which takes the above code path. There's a comment following this that says "DNSKEYs were already validated by the inner query in the above lookup", but this is not the case. To fully verify the whole RRset of DNSKEYs, it would be necessary to check self-signatures by the trusted key over the other keys. Later in verify_default_rrset(), verify_rrset_with_dnskey() is called multiple times with different keys and signatures, and if any call succeeds, then its Proof is returned.

Similarly, verify_dnskey_rrset() returns Ok(false) if any DNSKEY record is covered by a DS record. A comment says "If all the keys are valid, then we are secure", but this is only checking that one key is authenticated by a DS in the parent zone's delegation point. This time, after control flow returns to verify_rrset(), it will call verify_default_rrset(). The special handling for DNSKEYs in verify_default_rrset() will then call verify_rrset_with_dnskey() using each KSK DNSKEY record, and if one call succeeds, return its Proof. If there are multiple KSK DNSKEYs in the RRset, then this leads to another authentication break. We need to either pass the authenticated DNSKEYs from the DS covering check to the RRSIG validation, or we need to perform this RRSIG validation of the DNSKEY RRset inside verify_dnskey_rrset() and cut verify_default_rrset() out of DNSKEY RRset validation entirely.

ring::aead::quic::HeaderProtectionKey::new_mask() may panic when overflow checking is enabled. In the QUIC protocol, an attacker can induce this panic by sending a specially-crafted packet. Even unintentionally it is likely to occur in 1 out of every 2**32 packets sent and/or received.

On 64-bit targets operations using ring::aead::{AES_128_GCM, AES_256_GCM} may panic when overflow checking is enabled, when encrypting/decrypting approximately 68,719,476,700 bytes (about 64 gigabytes) of data in a single chunk. Protocols like TLS and SSH are not affected by this because those protocols break large amounts of data into small chunks. Similarly, most applications will not attempt to encrypt/decrypt 64GB of data in one chunk.

Overflow checking is not enabled in release mode by default, but RUSTFLAGS="-C overflow-checks" or overflow-checks = true in the Cargo.toml profile can override this. Overflow checking is usually enabled by default in debug mode.

When a Some(...) value was passed to the properties argument of either of these functions, a use-after-free would result.

In practice this would nearly always result in OpenSSL treating the properties as an empty string (due to CString::drop's behavior).

The maintainers thank quitbug for reporting this vulnerability to us.

Previous versions of tracing-subscriber were vulnerable to ANSI escape sequence injection attacks. Untrusted user input containing ANSI escape sequences could be injected into terminal output when logged, potentially allowing attackers to:

• Manipulate terminal title bars
• Clear screens or modify terminal display
• Potentially mislead users through terminal manipulation

In isolation, impact is minimal, however security issues have been found in terminal emulators that enabled an attacker to use ANSI escape sequences via logs to exploit vulnerabilities in the terminal emulator.

This was patched in PR #3368 to escape ANSI control characters from user input.

maxminddb prior to version 0.27 declared Reader::open_mmap as safe despite wrapping an inherently unsafe memmap2 operation with no extra step done to guarantee safety. This could have led to undefined behaviour if the file were to be modified on disk while the memory map was still active.