Some servers which misunderstood the point of the CertificateRequest
message send huge reply records. These records are large enough that
they were considered “insane” by the TLS code and rejected.
This change removes the sanity test for record lengths. Although the
maxCiphertext test still remains, just above, which (roughly) enforces
the 16KB protocol limit on record sizes:
https://tools.ietf.org/html/rfc5246#section-6.2.1Fixes#8928.
Change-Id: Idf89a2561b1947325b7ddc2613dc2da638d7d1c9
Reviewed-on: https://go-review.googlesource.com/5690
Reviewed-by: Andrew Gerrand <adg@golang.org>
Reviewed-by: Brad Fitzpatrick <bradfitz@golang.org>
Update #3514
An io.Reader is permitted to return either (n, nil)
or (n, io.EOF) on EOF or other error.
The tls package previously always returned (n, nil) for a read
of size n if n bytes were available, not surfacing errors at
the same time.
Amazon's HTTPS frontends like to hang up on clients without
sending the appropriate HTTP headers. (In their defense,
they're allowed to hang up any time, but generally a server
hangs up after a bit of inactivity, not immediately.) In any
case, the Go HTTP client tries to re-use connections by
looking at whether the response headers say to keep the
connection open, and because the connection looks okay, under
heavy load it's possible we'll reuse it immediately, writing
the next request, just as the Transport's always-reading
goroutine returns from tls.Conn.Read and sees (0, io.EOF).
But because Amazon does send an AlertCloseNotify record before
it hangs up on us, and the tls package does its own internal
buffering (up to 1024 bytes) of pending data, we have the
AlertCloseNotify in an unread buffer when our Conn.Read (to
the HTTP Transport code) reads its final bit of data in the
HTTP response body.
This change makes that final Read return (n, io.EOF) when
an AlertCloseNotify record is buffered right after, if we'd
otherwise return (n, nil).
A dependent change in the HTTP code then notes whether a
client connection has seen an io.EOF and uses that as an
additional signal to not reuse a HTTPS connection. With both
changes, the majority of Amazon request failures go
away. Without either one, 10-20 goroutines hitting the S3 API
leads to such an error rate that empirically up to 5 retries
are needed to complete an API call.
LGTM=agl, rsc
R=agl, rsc
CC=golang-codereviews
https://golang.org/cl/76400046
Currently a write error will cause future reads to return that same error.
However, there may have been extra information from a peer pending on
the read direction that is now unavailable.
This change splits the single connErr into errors for the read, write and
handshake. (Splitting off the handshake error is needed because both read
and write paths check the handshake error.)
Fixes#7414.
LGTM=bradfitz, r
R=golang-codereviews, r, bradfitz
CC=golang-codereviews
https://golang.org/cl/69090044
Adam (agl@) had already done an initial review of this CL in a branch.
Added ClientSessionState to Config which now allows clients to keep state
required to resume a TLS session with a server. A client handshake will try
and use the SessionTicket/MasterSecret in this cached state if the server
acknowledged resumption.
We also added support to cache ClientSessionState object in Config that will
be looked up by server remote address during the handshake.
R=golang-codereviews, agl, rsc, agl, agl, bradfitz, mikioh.mikioh
CC=golang-codereviews
https://golang.org/cl/15680043
AES-GCM is the only current TLS ciphersuite that doesn't have
cryptographic weaknesses (RC4), nor major construction issues (CBC mode
ciphers) and has some deployment (i.e. not-CCM).
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/13249044
This does not include AES-GCM yet. Also, it assumes that the handshake and
certificate signature hash are always SHA-256, which is true of the ciphersuites
that we currently support.
R=golang-dev, rsc
CC=golang-dev
https://golang.org/cl/10762044
The significant change between TLS 1.0 and 1.1 is the addition of an explicit IV in the case of CBC encrypted records. Support for TLS 1.1 is needed in order to support TLS 1.2.
R=golang-dev, bradfitz
CC=golang-dev
https://golang.org/cl/7880043
OpenSSL can be configured to send empty records in order to randomise
the CBC IV. This is an early version of 1/n-1 record splitting (that Go
does) and is quite reasonable, but it results in tls.Conn.Read
returning (0, nil).
This change ignores up to 100 consecutive, empty records to avoid
returning (0, nil) to callers.
Fixes 5309.
R=golang-dev, r, minux.ma
CC=golang-dev
https://golang.org/cl/8852044
The RFC doesn't actually have an opinion on whether this is a fatal or
warning level alert, but common practice suggests that it should be a
warning.
This involves rebasing most of the tests.
Fixes#3413.
R=golang-dev, shanemhansen, rsc
CC=golang-dev
https://golang.org/cl/6654050
Session resumption saves a round trip and removes the need to perform
the public-key operations of a TLS handshake when both the client and
server support it (which is true of Firefox and Chrome, at least).
R=golang-dev, bradfitz, rsc
CC=golang-dev
https://golang.org/cl/6555051
Fixes#3862.
There were many areas where conn.err was being accessed
outside the mutex. This proposal moves the err value to
an embedded struct to make it more obvious when the error
value is being accessed.
As there are no Benchmark tests in this package I cannot
feel confident of the impact of this additional locking,
although most will be uncontended.
R=dvyukov, agl
CC=golang-dev
https://golang.org/cl/6497070
1) Remove the Reset() member in crypto/aes and crypto/des (and
document the change).
2) Turn several empty error structures into vars. Any remaining error
structures are either non-empty, or will probably become so in the
future.
3) Implement SetWriteDeadline for TLS sockets. At the moment, the TLS
status cannot be reused after a Write error, which is probably fine
for most uses.
4) Make crypto/aes and crypto/des return a cipher.Block.
R=rsc, r
CC=golang-dev
https://golang.org/cl/5625045
Previously, a timeout (in int64 nanoseconds) applied to a granularity
even smaller than one operation: a 100 byte read with a 1 second timeout
could take 100 seconds, if the bytes all arrived on the network 1 second
apart. This was confusing.
Rather than making the timeout granularity be per-Read/Write,
this CL makes callers set an absolute deadline (in time.Time)
after which operations will fail. This makes it possible to
set deadlines at higher levels, without knowing exactly how
many read/write operations will happen in e.g. reading an HTTP
request.
Fixes#2723
R=r, rsc, dave
CC=golang-dev
https://golang.org/cl/5555048
The code in hash functions themselves could write directly into the
output buffer for a savings of about 50ns. But it's a little ugly so I
wasted a copy.
R=bradfitz
CC=golang-dev
https://golang.org/cl/5440111
tls.Conn.Close() didn't close the underlying connection and tried to
do a handshake in order to send the close notify alert.
http didn't look for errors from the TLS handshake.
Fixes#2281.
R=bradfitz
CC=golang-dev
https://golang.org/cl/5283045
It would be nice not to have to support this since all the clients
that we care about support TLSv1 by now. However, due to buggy
implementations of SSLv3 on the Internet which can't do version
negotiation correctly, browsers will sometimes switch to SSLv3. Since
there's no good way for a browser tell a network problem from a buggy
server, this downgrade can occur even if the server in question is
actually working correctly.
So we need to support SSLv3 for robustness :(
Fixes#1703.
R=bradfitz
CC=golang-dev
https://golang.org/cl/5018045
This is a core API change.
1) gofix misc src
2) Manual adjustments to the following files under src/pkg:
gob/decode.go
rpc/client.go
os/error.go
io/io.go
bufio/bufio.go
http/request.go
websocket/client.go
as well as:
src/cmd/gofix/testdata/*.go.in (reverted)
test/fixedbugs/bug243.go
3) Implemented gofix patch (oserrorstring.go) and test case (oserrorstring_test.go)
Compiles and runs all tests.
R=r, rsc, gri
CC=golang-dev
https://golang.org/cl/4607052
People have a need to verify certificates in situations other than TLS
client handshaking. Thus this CL moves certificate verification into
x509 and expands its abilities.
R=bradfitzgo
CC=golang-dev
https://golang.org/cl/4407046
This is largely based on ality's CL 2747042.
crypto/rc4: API break in order to conform to crypto/cipher's
Stream interface
cipher/cipher: promote to the default build
Since CBC differs between TLS 1.0 and 1.1, we downgrade and
support only 1.0 at the current time. 1.0 is what most of the
world uses.
Given this CL, it would be trival to add support for AES 256,
SHA 256 etc, but I haven't in order to keep the change smaller.
R=rsc
CC=ality, golang-dev
https://golang.org/cl/3659041
Changed all uses of bytes.Add (aside from those testing bytes.Add) to append(a, b...).
Also ran "gofmt -s" and made use of copy([]byte, string) in the fasta benchmark.
R=golang-dev, r, r2
CC=golang-dev
https://golang.org/cl/3302042
asn1: add support for T61String because this is the string type which
several www.google.com certificates are now using for fields like
CommonName
tls: force a handshake in Dial so that certificates are ready
afterwards.
Fixes#1114.
R=rsc
CC=golang-dev
https://golang.org/cl/2216043
This changeset implements client certificate support in crypto/tls
for both handshake_server.go and handshake_client.go
The updated server implementation sends an empty CertificateAuthorities
field in the CertificateRequest, thus allowing clients to send any
certificates they wish. Likewise, the client code will only respond
with its certificate when the server requests a certificate with this
field empty.
R=agl, rsc, agl1
CC=golang-dev
https://golang.org/cl/1975042
SNI (Server Name Indication) is a way for a TLS client to
indicate to the server which name it knows the server by. This
allows the server to have several names and return the correct
certificate for each (virtual hosting).
PeerCertificates returns the list of certificates presented by
server.
R=r
CC=golang-dev
https://golang.org/cl/1741053
