Tor before 0.2.0.34 treats incomplete IPv4 addresses as valid, which has unknown impact and attack vectors related to "Spec conformance," as demonstrated using 192.168.0.

Unspecified vulnerability in Tor before 0.2.0.33 has unspecified impact and remote attack vectors that trigger heap corruption.

Heap-based buffer overflow in Tor before 0.2.1.28 and 0.2.2.x before 0.2.2.20-alpha allows remote attackers to cause a denial of service (daemon crash) or possibly execute arbitrary code via unspecified vectors.

Tor before 0.2.0.32 does not properly process the ClientDNSRejectInternalAddresses configuration option in situations where an exit relay issues a policy-based refusal of a stream, which allows remote exit relays to have an unknown impact by mapping an internal IP address to the destination hostname of a refused stream.

Multiple heap-based buffer overflows in Tor before 0.2.2.35 allow remote attackers to cause a denial of service (memory corruption) or possibly execute arbitrary code by (1) establishing a SOCKS connection to SocksPort or (2) leveraging a SOCKS proxy configuration.

Integer overflow in Tor before 0.1.1.20 allows remote attackers to execute arbitrary code via crafted large inputs, which result in a buffer overflow when elements are added to smartlists.

debian/tor.init in the Debian tor0.2.9.11-1~deb9u1 package for Tor was designed to execute aa-exec from the standard system pathname if the apparmor package is installed, but implements this incorrectly (with a wrong assumption that the specific pathname would remain the same forever), which allows attackers to bypass intended AppArmor restrictions by leveraging the silent loss of this protection mechanism. NOTE: this does not affect systems, such as default Debian stretch installations, on which Tor startup relies on a systemd unit file (instead of this tor.init script).

Tor before 0.2.0.32 does not properly process the (1) User and (2) Group configuration options, which might allow local users to gain privileges by leveraging unintended supplementary group memberships of the Tor process.

Heap-based buffer overflow in Tor before 0.2.1.29 and 0.2.2.x before 0.2.2.21-alpha allows remote attackers to cause a denial of service (memory corruption and application crash) or possibly execute arbitrary code via unspecified vectors.

TLS handshakes in Tor before 0.1.1.20 generate public-private keys based on TLS context rather than the connection, which makes it easier for remote attackers to conduct brute force attacks on the encryption keys.

Tor before 0.1.1.20 allows remote attackers to spoof log entries or possibly execute shell code via strings with non-printable characters.

Tor before 0.1.1.20 does not sufficiently obey certain firewall options, which allows remote attackers to bypass intended access restrictions for dirservers, direct connections, or proxy servers.

Tor before 0.1.1.20 uses improper logic to validate the "OR" destination, which allows remote attackers to perform a man-in-the-middle (MITM) attack via unspecified vectors.

Tor client before 0.1.1.20 prefers entry points based on isfast or isstable flags, which could allow remote attackers to be preferred over nodes that are identified as more trustworthy "entry guard" (isguard) systems by directory authorities.

Tor before 0.1.2.15 sends "destroy cells" containing the reason for tearing down a circuit, which allows remote attackers to obtain sensitive information, contrary to specifications.

Tor before 0.1.2.16, when ControlPort is enabled, does not properly restrict commands to localhost port 9051, which allows remote attackers to modify the torrc configuration file, compromise anonymity, and have other unspecified impact via HTTP POST data containing commands without valid authentication, as demonstrated by an HTML form (1) hosted on a web site or (2) injected by a Tor exit node.

Tor before 0.1.2.15 can select a guard node beyond the first listed never-before-connected-to guard node, which allows remote attackers with control of certain guard nodes to obtain sensitive information and possibly leverage further attacks.

Tor before 0.1.2.15 does not properly distinguish "streamids from different exits," which might allow remote attackers with control over Tor routers to inject cells into arbitrary streams.

Buffer overflow in Tor before 0.1.2.15, when using BSD natd support, allows remote attackers to cause a denial of service via unspecified vectors.

Tor before 0.2.2.34, when configured as a client or bridge, sends a TLS certificate chain as part of an outgoing OR connection, which allows remote relays to bypass intended anonymity properties by reading this chain and then determining the set of entry guards that the client or bridge had selected.

Tor 0.2.0.28, and probably 0.2.0.34 and earlier, allows remote attackers, with control of an entry router and an exit router, to confirm that a sender and receiver are communicating via vectors involving (1) replaying, (2) modifying, (3) inserting, or (4) deleting a single cell, and then observing cell recognition errors at the exit router. NOTE: the vendor disputes the significance of this issue, noting that the product's design "accepted end-to-end correlation as an attack that is too expensive to solve."

Tor 0.1.0.13 and earlier, and experimental versions 0.1.1.4-alpha and earlier, does not reject certain weak keys when using ephemeral Diffie-Hellman (DH) handshakes, which allows malicious Tor servers to obtain the keys that a client uses for other systems in the circuit.

Unknown vulnerability in Tor before 0.1.0.10 allows remote attackers to read arbitrary memory and possibly key information from the exit server's process space.

Tor before 0.1.1.20 allows remote attackers to identify hidden services via a malicious Tor server that attempts a large number of accesses of the hidden service, which eventually causes a circuit to be built through the malicious server.

Unspecified vulnerability in the directory server (dirserver) in Tor before 0.1.1.20 allows remote attackers to cause an unspecified denial of service via unknown vectors.

Tor before 0.1.1.20 uses OpenSSL pseudo-random bytes (RANDpseudobytes) instead of cryptographically strong RANDbytes, and seeds the entropy value at start-up with 160-bit chunks without reseeding, which makes it easier for attackers to conduct brute force guessing attacks.

The privoxy configuration file in Tor before 0.1.1.20, when run on Apple OS X, logs all data via the "logfile", which allows attackers to obtain potentially sensitive information.

Tor before 0.1.1.20 does not validate that a server descriptor's fingerprint line matches its identity key, which allows remote attackers to spoof the fingerprint line, which might be trusted by users or other applications.

DISPUTED Tor before 0.1.1.20 kills the circuit when it receives an unrecognized relay command, which causes network circuits to be disbanded. NOTE: while this item is listed under the "Security fixes" section of the developer changelog, the developer clarified on 20060707 that this is only a self-DoS. Therefore this issue should not be included in CVE.

Tor before 0.1.1.20 supports server descriptors that contain hostnames instead of IP addresses, which allows remote attackers to arbitrarily group users by providing preferential address resolution.