In the SSH protocol, clients and servers execute the key exchange to
generate one-time session keys used for encryption and authentication.
The key exchange is performed initially after the connection is
established and then periodically after a configurable amount of data.
While a key exchange is in progress, we add the received packets to an
internal queue until we receive SSH_MSG_KEXINIT from the other side.
This can result in high memory usage if the other party is slow to
respond to the SSH_MSG_KEXINIT packet, or memory exhaustion if a
malicious client never responds to an SSH_MSG_KEXINIT packet during a
large file transfer.
We now limit the internal queue to 64 packets: this means 2MB with the
typical 32KB packet size.
When the internal queue is full we block further writes until the
pending key exchange is completed or there is a read or write error.
Thanks to Yuichi Watanabe for reporting this issue.
Change-Id: I1ce2214cc16e08b838d4bc346c74c72addafaeec
Reviewed-on: https://go-review.googlesource.com/c/crypto/+/652135
Reviewed-by: Neal Patel <nealpatel@google.com>
Auto-Submit: Gopher Robot <gobot@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
Implement the "strict KEX" protocol changes, as described in section
1.9 of the OpenSSH PROTOCOL file (as of OpenSSH version 9.6/9.6p1).
Namely this makes the following changes:
* Both the server and the client add an additional algorithm to the
initial KEXINIT message, indicating support for the strict KEX mode.
* When one side of the connection sees the strict KEX extension
algorithm, the strict KEX mode is enabled for messages originating
from the other side of the connection. If the sequence number for
the side which requested the extension is not 1 (indicating that it
has already received non-KEXINIT packets), the connection is
terminated.
* When strict kex mode is enabled, unexpected messages during the
handshake are considered fatal. Additionally when a key change
occurs (on the receipt of the NEWKEYS message) the message sequence
numbers are reset.
Thanks to Fabian Bäumer, Marcus Brinkmann, and Jörg Schwenk from Ruhr
University Bochum for reporting this issue.
Fixes CVE-2023-48795
Fixesgolang/go#64784
Change-Id: I96b53afd2bd2fb94d2b6f2a46a5dacf325357604
Reviewed-on: https://go-review.googlesource.com/c/crypto/+/550715
Reviewed-by: Nicola Murino <nicola.murino@gmail.com>
Reviewed-by: Tatiana Bradley <tatianabradley@google.com>
TryBot-Result: Gopher Robot <gobot@golang.org>
Run-TryBot: Roland Shoemaker <roland@golang.org>
Reviewed-by: Damien Neil <dneil@google.com>
LUCI-TryBot-Result: Go LUCI <golang-scoped@luci-project-accounts.iam.gserviceaccount.com>
The server implementation looks at the HostKeys to advertise and
negotiate host key signature algorithms. A fundamental issue of the
Signer and AlgorithmSigner interfaces is that they don't expose the
supported signature algorithms, so really the server has to guess.
Currently, it would guess exclusively based on the PublicKey.Type,
regardless of whether the host key implemented AlgorithmSigner. This
means that a legacy Signer that only supports ssh-rsa still led the
server to negotiate rsa-sha2 algorithms. The server would then fail to
find a suitable host key to make the signature and crash.
This won't happen if only Signers from this package are used, but if a
custom Signer that doesn't support SignWithAlgorithm() but returns
"ssh-rsa" from PublicKey().Type() is used as a HostKey, the server is
vulnerable to DoS.
The only workable rules to determine what to advertise seems to be:
1. a pure Signer will always Sign with the PublicKey.Type
2. an AlgorithmSigner supports all algorithms associated with the
PublicKey.Type
Rule number two means that we can't add new supported algorithms in the
future, which is not great, but it's too late to fix that.
rsaSigner was breaking rule number one, and although it would have been
fine where it's used, I didn't want to break our own interface contract.
It's unclear why we had separate test key entries for rsa-sha2
algorithms, since we can use the ssh-rsa key for those. The only test
that used them, TestCertTypes, seemed broken: the init was actually
failing at making the corresponding signers rsaSigners, and indeed the
test for the SHA-256 signer expected and checked a SHA-512 signature.
Pending CVE
For golang/go#49952
Change-Id: Ie658eefcadd87906e63fc7faae8249376aa96c79
Reviewed-on: https://go-review.googlesource.com/c/crypto/+/392355
Trust: Filippo Valsorda <filippo@golang.org>
Run-TryBot: Filippo Valsorda <filippo@golang.org>
Reviewed-by: Roland Shoemaker <roland@golang.org>
TryBot-Result: Gopher Robot <gobot@golang.org>
Dialing the 0.0.0.0 address (as returned by net.Addr().String() for a
net.Listen("tcp", ":1") address) is not yet guaranteed to work. It's
currently OS-dependent.
For some reason it works on Go 1.8+, but it hasn't yet been defined to
work reliably.
Fix the tests for now (since we need to support older Go releases),
even if this might work in the future.
Updates golang/go#18806
Change-Id: I2f0476b1d4f2673ab64ffedfa733f2d92fceb6ff
Reviewed-on: https://go-review.googlesource.com/42496
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
This change breaks existing behavior.
Before, a missing ClientConfig.HostKeyCallback would cause host key
checking to be disabled. In this configuration, establishing a
connection to any host just works, so today, most SSH client code in
the wild does not perform any host key checks.
This makes it easy to perform a MITM attack:
* SSH installations that use keyboard-interactive or password
authentication can be attacked with MITM, thereby stealing
passwords.
* Clients that use public-key authentication with agent forwarding are
also vulnerable: the MITM server could allow the login to succeed, and
then immediately ask the agent to authenticate the login to the real
server.
* Clients that use public-key authentication without agent forwarding
are harder to attack unnoticedly: an attacker cannot authenticate the
login to the real server, so it cannot in general present a convincing
server to the victim.
Now, a missing HostKeyCallback will cause the handshake to fail. This
change also provides InsecureIgnoreHostKey() and FixedHostKey(key) as
ready made host checkers.
A simplistic parser for OpenSSH's known_hosts file is given as an
example. This change does not provide a full-fledged parser, as it
has complexity (wildcards, revocation, hashed addresses) that will
need further consideration.
When introduced, the host checking feature maintained backward
compatibility at the expense of security. We have decided this is not
the right tradeoff for the SSH library.
Fixesgolang/go#19767
Change-Id: I45fc7ba9bd1ea29c31ec23f115cdbab99913e814
Reviewed-on: https://go-review.googlesource.com/38701
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
The ssh tests currently require 127.0.0.1 to work which isn't
necessarily available everywhere. To fix the source-address tests,
support comma-separated source-address values per the PROTOCOL.certkeys
file:
Comma-separated list of source addresses
from which this certificate is accepted
for authentication. Addresses are
specified in CIDR format (nn.nn.nn.nn/nn
or hhhh::hhhh/nn).
If this option is not present then
certificates may be presented from any
source address.
Change-Id: I87536ff81ffa005c073da103021ebc0dfb12b620
Reviewed-on: https://go-review.googlesource.com/36110
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Run-TryBot: Heschi Kreinick <heschi@google.com>
Since encryption messes up the packets, the wrongly retained packets
look like noise and cause application protocol errors or panics in the
SSH library.
This normally triggers very rarely: the mandatory key exchange doesn't
have parallel writes, so this failure condition would be setup on the
first key exchange, take effect only after the second key exchange.
Fortunately, the tests against openssh exercise this. This change adds
also adds a unittest.
Fixes#18850.
Change-Id: I656c8b94bfb265831daa118f4d614a2f0c65d2af
Reviewed-on: https://go-review.googlesource.com/36056
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Use channels and a dedicated write loop for managing the rekeying
process. This lets us collect packets to be written while a key
exchange is in progress.
Previously, the read loop ran the key exchange, and writers would
block if a key exchange was going on. If a reader wrote back a packet
while processing a read packet, it could block, stopping the read
loop, thus causing a deadlock. Such coupled read/writes are inherent
with handling requests that want a response (eg. keepalive,
opening/closing channels etc.). The buffered channels (most channels
have capacity 16) papered over these problems, but under load SSH
connections would occasionally deadlock.
Fixes#18439.
Change-Id: I7c14ff4991fa3100a5d36025125d0cf1119c471d
Reviewed-on: https://go-review.googlesource.com/35012
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
This is done by running the key exchange and setting the session ID
under mutex. If the first exchange encounters an already set session
ID, then do nothing.
This fixes a race condition:
On setting up the connection, both sides sent a kexInit to initiate
the first (mandatory) key exchange. If one side was faster, the
faster side might have completed the key exchange, before the slow
side had a chance to send a kexInit. The slow side would send a
kexInit which would trigger a second key exchange. The resulting
confirmation message (msgNewKeys) would confuse the authentication
loop.
This fix removes sessionID from the transport struct.
This fix also deletes the unused interface rekeyingTransport.
Fixes#15066
Change-Id: I7f303bce5d3214c9bdd58f52d21178a185871d90
Reviewed-on: https://go-review.googlesource.com/21606
Reviewed-by: Adam Langley <agl@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Unblock writers if a read error occurs while writers are blocked on a
pending key change.
Add test to check for deadlocks in error paths in handshake.go
Fixesgolang/go#11992.
Change-Id: Id098bd9fec3d4fe83daeb2b7f935e5647c19afd3
Reviewed-on: https://go-review.googlesource.com/13594
Reviewed-by: Adam Langley <agl@golang.org>
cuishuang
Nicola Murino
Roland Shoemaker
Tobias Klauser
Filippo Valsorda
Roland Shoemaker
Agniva De Sarker
Brad Fitzpatrick
Han-Wen Nienhuys
Heschi Kreinick
David du Colombier
Adam Langley