Forgotten file from CL 112635.
The integration tests didn't catch this because the integration tests
never run (sshd isn't in $PATH usually), but even when $PATH is
modified, it seems to have been failing for ages on both Debian jessie
and stretch. I'll fix the tests and make them run in TryBots in later
changes.
Change-Id: I85fd3b6109a73990bc353a61da5f527b9a698501
Reviewed-on: https://go-review.googlesource.com/113056
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The extra goroutines were distracting while debugging something else,
especially as I wasn't using that feature. This also saves a bit of
memory.
Change-Id: Ia6489e64bbd3d5a6ff699a25018676d8ff8bd2e4
Reviewed-on: https://go-review.googlesource.com/112635
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
The tag matches the platforms defined in util*.go
where the most tested logic is defined.
Change-Id: I90f67d988c795738c3effbc8554a933a7cb355d2
Reviewed-on: https://go-review.googlesource.com/112555
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Doesn't matter because contentKey and lengthKey are the same length,
but we should really be looping over lengthKey here.
Change-Id: Ic7af3c553eb4382239bec1c030e77c7f5baa0662
Reviewed-on: https://go-review.googlesource.com/109735
Run-TryBot: Michael Munday <mike.munday@ibm.com>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Unlock compares the length of the passphrase with the given one before
calling subtle.ConstantTimeCompare. This is redundant, since
ConstantTimeCompare already perform a lengths check before doing
anything. Remove the check from Unlock.
Updates golang/go#25173
Change-Id: Ib5fec3a94392bddf2996f5c6bf5a414529e86f2f
Reviewed-on: https://go-review.googlesource.com/110068
Run-TryBot: Alberto Donizetti <alb.donizetti@gmail.com>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
This fixes the following vulnerability scenario:
* Victim logs into SAFE-HOST on SAFE-IP-ADDRESS regularly.
* Victim is cajoled into connecting to attacker controlled
ATTACK-HOST, on ATTACK-IP-ADDRESS. ATTACK-HOST uses a different host
key type (e.g. Ed25519 vs RSA). The new key is added at the end of
known_hosts.
* Attacker makes DNS system return ATTACK-IP-ADDRESS for SAFE-HOST.
* Victim logs into SAFE-HOST, but is not warned because the host key
matches ATTACK-IP-ADDRESS.
For this attack to work, the key type has to be different, because
knownhosts gives precedence to the first key found for each type. Add
a test that asserts this behavior.
The new semantics simplify the code, but callers that modify
.ssh/known_host interactviely must now take an extra step to remain
OpenSSH compatible: on successful login, the IP address must be
checked without hostname, and if it is not known, added separately to
the known_hosts file, so future logins that use an IP address only
will be protected too.
Thanks to Daniel Parks <security@demonhorse.net> for finding this
vulnerability.
Change-Id: I62b1b60ceb02e2f583a4657213feac1a8885dd42
Reviewed-on: https://go-review.googlesource.com/104939
Reviewed-by: Adam Langley <agl@golang.org>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The extra Close call runs in a separate goroutine, and races with the
deferred Close call, leading to test flakes if the extra call fails
Change-Id: Ia1a2fb3932906da7cd0a8854b55f82785bb33716
Reviewed-on: https://go-review.googlesource.com/104335
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
TestMakeRawState currently fails on Solaris:
--- FAIL: TestMakeRawState (0.00s)
terminal_test.go:334: states do not match; was &{0xc420015dd0}, expected &{0xc420015da0}
Change terminal.State to include a copy to the unix.Termios (like the
implementation for Linux and the BSDs) which also makes terminal.MakeRaw behave
as expected and lets TestMakeRawState pass.
Change-Id: I29382f83b84ff301991e1db170f32f41e144aec8
Reviewed-on: https://go-review.googlesource.com/99456
Run-TryBot: Tobias Klauser <tobias.klauser@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
yet.
Since ServerAuthError returns all errors collected during the
userAuthLoop, it will always include the generic error "no auth passed
yet" (assuming the connection fails).
Change-Id: I5f6c67f3f0762b023618178d4028600d2b6c9253
Reviewed-on: https://go-review.googlesource.com/92737
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
The existing client side authentication does not handle correctly
the partial success flag in SSH_MSG_USERAUTH_FAILURE authentication
responses.
This commit fixes two problems in ssh library:
1) RetryableAuthMethod() now breaks out from the retry loop and
returns when underlying auth method fails with partial success
set to true.
2) Book keeping of tried (and failed) auth methods in
clientAuthenticate() does not mark an auth method failed if it
fails with partial success set to true.
Fixesgolang/go#23461
Change-Id: Ib2e1a1d54bfe2549496199bb2f66ebbce58d130d
Reviewed-on: https://go-review.googlesource.com/88035
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
os.NewFile assigns finalizer to close file handle
passed into ReadPassword. But that is not expected.
Make a duplicate of original file handle, and pass
copy handle into ReadPassword instead.
Fixesgolang/go#23525
Change-Id: I4d6725e9a1cc20defd1b58afc383e35a7f9ee4e9
Reviewed-on: https://go-review.googlesource.com/89395
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Incomplete reads leave (part of) the verification tag zeroed, leading
to a failing MAC, and this is more likely to happen with larger
packets. The test added in the previous commit should stop this from
regressing.
Reinstate chacha20-poly1305 as a default cipher and prefer it over AES
CTR flavors.
Fixesgolang/go#23510
Change-Id: I7599897e59448edb7b814eebcc8226ea15b365d6
Reviewed-on: https://go-review.googlesource.com/89075
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
This effectively disables the cipher.
Add a test against OpenSSH which sends larger packets through the
tested ciphers. This reproduces the problem reported in golang/go#23510
Change-Id: I4b124c690c409c6a0af2621e58a964ff55815f57
Reviewed-on: https://go-review.googlesource.com/88995
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Adam Langley <agl@golang.org>
Fixesgolang/go#23194
During SSH Protocol Version Exchange, a client may send metadata lines
prior to sending the SSH version string. To conform to the RFC, all SSH
implementations must support this (minimally, clients can ignore the
metadata lines).
For example, this is valid:
some-metadata
SSH-2.0-OpenSSH
The current Go implementation takes the first line it sees as
the version string (in this case, some-metadata). Then, it uses
the next line (SSH-2.0-OpenSSH) as part of key exchange, which
is guaranteed to fail.
Unfortunately, this SSH feature is used by some vendors and is part
of the official RFC: https://tools.ietf.org/html/rfc4253#section-4.2
Change-Id: I7be61700a07756353875bf43aad09a580ba533ff
Reviewed-on: https://go-review.googlesource.com/86675
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Per the description, the "-s" file issues slight simplifications to
the source code.
Change-Id: I77395d763f5eafb48653902dcedfa56b150b2d67
Reviewed-on: https://go-review.googlesource.com/80138
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
None are "wrong" per se, but there are a lot of good suggestions and
in one case a docstring that was not present in godoc due to the
presence of an extra newline.
Changed "Id" in struct properties to "ID" in some non-exported
structs. Removed a trailing period from some error messages; I believe
the exact contents of error strings are not covered by the Go
compatibility promise.
Change-Id: I7c620582dc247396f72c52d38c909ccc0ec87b83
Reviewed-on: https://go-review.googlesource.com/80145
Run-TryBot: Brad Fitzpatrick <bradfitz@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Only show the SSH banner once, even if the client attempts
authentication with the "none" type and resets the authentication
attempts.
Change-Id: I1a7aacb50abf00233ed4d06c60808aaf51a9d7be
Reviewed-on: https://go-review.googlesource.com/78544
Run-TryBot: Sam Whited <sam@samwhited.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
ReadPassword uses Windows ReadFile to read from console handle.
But ReadFile does not split input into UTF-8 characters, so ReadFile
only works when input is ASCII. Use os.File instead of Windows
ReadFile, because os.File reads console and parses it into UTF-8.
Fixesgolang/go#22828
Change-Id: Ifeed3e8048b51f46706c28d4154a3e4b10111a3e
Reviewed-on: https://go-review.googlesource.com/79335
Reviewed-by: Alex Brainman <alex.brainman@gmail.com>
Run-TryBot: Alex Brainman <alex.brainman@gmail.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
According to RFC 4252 section 5.4, the banner is sent between the
ssh-connection request and responding to user authentication.
Original support for server sending banner by joshua stein <jcs@jcs.org>
Fixesgolang/go#19567
Change-Id: I729b3c8e5fd2c0068609d1590b61e92f40d87ea4
Reviewed-on: https://go-review.googlesource.com/71790
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
According to RFC 4252 section 5.4, the banner is sent between the
ssh-connection request and responding to user authentication.
Original support for server sending banner by joshua stein <jcs@jcs.org>
Fixesgolang/go#19567
Change-Id: I68944a7f4711c0623759f6a59023e8e45a8781aa
Reviewed-on: https://go-review.googlesource.com/65271
Reviewed-by: Han-Wen Nienhuys <hanwen@google.com>
Run-TryBot: Han-Wen Nienhuys <hanwen@google.com>
The Solaris version of MakeRaw already sets VMIN and VTIME explicitly
such that a read returns when one character is available.
cfmakeraw (whose behavior MakeRaw replicate) in glibc and the BSD's libc
also set these values explicitly, so it should be done in the Linux/BSD
versions of MakeRaw as well to be consistent.
Change-Id: I531641ec87fd6a21b7a544b9a464bb90045b0bb1
Reviewed-on: https://go-review.googlesource.com/53570
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Avelino <t@avelino.xxx>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Noel Georgi
Tobias Klauser
Brad Fitzpatrick
Hana (Hyang-Ah) Kim
Michael Munday
Alberto Donizetti
Agniva De Sarker
Han-Wen Nienhuys
Spencer Tung
Sami Pönkänen
Sergey Lukjanov
Yasuhiro Matsumoto
Justin Gracenin
Mansour Rahimi
Peter Moody
Kevin Burke
Sam Whited
Tugdual Saunier
namusyaka
Gh0u1L5
Artyom Pervukhin