New cfilter HTTP-CONNECT for h3/h2/http1.1 eyeballing.
- filter is installed when `--http3` in the tool is used (or
the equivalent CURLOPT_ done in the library)
- starts a QUIC/HTTP/3 connect right away. Should that not
succeed after 100ms (subject to change), a parallel attempt
is started for HTTP/2 and HTTP/1.1 via TCP
- both attempts are subject to IPv6/IPv4 eyeballing, same
as happens for other connections
- tie timeout to the ip-version HAPPY_EYEBALLS_TIMEOUT
- use a `soft` timeout at half the value. When the soft timeout
expires, the HTTPS-CONNECT filter checks if the QUIC filter
has received any data from the server. If not, it will start
the HTTP/2 attempt.
HTTP/3(ngtcp2) improvements.
- setting call_data in all cfilter calls similar to http/2 and vtls filters
for use in callback where no stream data is available.
- returning CURLE_PARTIAL_FILE for prematurely terminated transfers
- enabling pytest test_05 for h3
- shifting functionality to "connect" UDP sockets from ngtcp2
implementation into the udp socket cfilter. Because unconnected
UDP sockets are weird. For example they error when adding to a
pollset.
HTTP/3(quiche) improvements.
- fixed upload bug in quiche implementation, now passes 251 and pytest
- error codes on stream RESET
- improved debug logs
- handling of DRAIN during connect
- limiting pending event queue
HTTP/2 cfilter improvements.
- use LOG_CF macros for dynamic logging in debug build
- fix CURLcode on RST streams to be CURLE_PARTIAL_FILE
- enable pytest test_05 for h2
- fix upload pytests and improve parallel transfer performance.
GOAWAY handling for ngtcp2/quiche
- during connect, when the remote server refuses to accept new connections
and closes immediately (so the local conn goes into DRAIN phase), the
connection is torn down and a another attempt is made after a short grace
period.
This is the behaviour observed with nghttpx when we tell it to shut
down gracefully. Tested in pytest test_03_02.
TLS improvements
- ALPN selection for SSL/SSL-PROXY filters in one vtls set of functions, replaces
copy of logic in all tls backends.
- standardized the infof logging of offered ALPNs
- ALPN negotiated: have common function for all backends that sets alpn proprty
and connection related things based on the negotiated protocol (or lack thereof).
- new tests/tests-httpd/scorecard.py for testing h3/h2 protocol implementation.
Invoke:
python3 tests/tests-httpd/scorecard.py --help
for usage.
Improvements on gathering connect statistics and socket access.
- new CF_CTRL_CONN_REPORT_STATS cfilter control for having cfilters
report connection statistics. This is triggered when the connection
has completely connected.
- new void Curl_pgrsTimeWas(..) method to report a timer update with
a timestamp of when it happend. This allows for updating timers
"later", e.g. a connect statistic after full connectivity has been
reached.
- in case of HTTP eyeballing, the previous changes will update
statistics only from the filter chain that "won" the eyeballing.
- new cfilter query CF_QUERY_SOCKET for retrieving the socket used
by a filter chain.
Added methods Curl_conn_cf_get_socket() and Curl_conn_get_socket()
for convenient use of this query.
- Change VTLS backend to query their sub-filters for the socket when
checks during the handshake are made.
HTTP/3 documentation on how https eyeballing works.
TLS improvements
- ALPN selection for SSL/SSL-PROXY filters in one vtls set of functions, replaces
copy of logic in all tls backends.
- standardized the infof logging of offered ALPNs
- ALPN negotiated: have common function for all backends that sets alpn proprty
and connection related things based on the negotiated protocol (or lack thereof).
Scorecard with Caddy.
- configure can be run with `--with-test-caddy=path` to specify which caddy to use for testing
- tests/tests-httpd/scorecard.py now measures download speeds with caddy
pytest improvements
- adding Makfile to clean gen dir
- adding nghttpx rundir creation on start
- checking httpd version 2.4.55 for test_05 cases where it is needed. Skipping with message if too old.
- catch exception when checking for caddy existance on system.
Closes#10349
- new functions and macros for cfilter debugging
- set CURL_DEBUG with names of cfilters where debug logging should be
enabled
- use GNUC __attribute__ to enable printf format checks during compile
Closes#10271
- they are mostly pointless in all major jurisdictions
- many big corporations and projects already don't use them
- saves us from pointless churn
- git keeps history for us
- the year range is kept in COPYING
checksrc is updated to allow non-year using copyright statements
Closes#10205
- almost all backend calls pass the Curl_cfilter intance instead of
connectdata+sockindex
- ssl_connect_data is remove from struct connectdata and made internal
to vtls
- ssl_connect_data is allocated in the added filter, kept at cf->ctx
- added function to let a ssl filter access its ssl_primary_config and
ssl_config_data this selects the propert subfields in conn and data,
for filters added as plain or proxy
- adjusted all backends to use the changed api
- adjusted all backends to access config data via the exposed
functions, no longer using conn or data directly
cfilter renames for clear purpose:
- methods `Curl_conn_*(data, conn, sockindex)` work on the complete
filter chain at `sockindex` and connection `conn`.
- methods `Curl_cf_*(cf, ...)` work on a specific Curl_cfilter
instance.
- methods `Curl_conn_cf()` work on/with filter instances at a
connection.
- rebased and resolved some naming conflicts
- hostname validation (und session lookup) on SECONDARY use the same
name as on FIRST (again).
new debug macros and removing connectdata from function signatures where not
needed.
adapting schannel for new Curl_read_plain paramter.
Closes#9919
- general construct/destroy in connectdata
- default implementations of callback functions
- connect: cfilters for connect and accept
- socks: cfilter for socks proxying
- http_proxy: cfilter for http proxy tunneling
- vtls: cfilters for primary and proxy ssl
- change in general handling of data/conn
- Curl_cfilter_setup() sets up filter chain based on data settings,
if none are installed by the protocol handler setup
- Curl_cfilter_connect() boot straps filters into `connected` status,
used by handlers and multi to reach further stages
- Curl_cfilter_is_connected() to check if a conn is connected,
e.g. all filters have done their work
- Curl_cfilter_get_select_socks() gets the sockets and READ/WRITE
indicators for multi select to work
- Curl_cfilter_data_pending() asks filters if the have incoming
data pending for recv
- Curl_cfilter_recv()/Curl_cfilter_send are the general callbacks
installed in conn->recv/conn->send for io handling
- Curl_cfilter_attach_data()/Curl_cfilter_detach_data() inform filters
and addition/removal of a `data` from their connection
- adding vtl functions to prevent use of Curl_ssl globals directly
in other parts of the code.
Reviewed-by: Daniel Stenberg
Closes#9855
Next Protocol Negotiation is a TLS extension that was created and used
for agreeing to use the SPDY protocol (the precursor to HTTP/2) for
HTTPS. In the early days of HTTP/2, before the spec was finalized and
shipped, the protocol could be enabled using this extension with some
servers.
curl supports the NPN extension with some TLS backends since then, with
a command line option `--npn` and in libcurl with
`CURLOPT_SSL_ENABLE_NPN`.
HTTP/2 proper is made to use the ALPN (Application-Layer Protocol
Negotiation) extension and the NPN extension has no purposes
anymore. The HTTP/2 spec was published in May 2015.
Today, use of NPN in the wild should be extremely rare and most likely
totally extinct. Chrome removed NPN support in Chrome 51, shipped in
June 2016. Removed in Firefox 53, April 2017.
Closes#9307
Add licensing and copyright information for all files in this repository. This
either happens in the file itself as a comment header or in the file
`.reuse/dep5`.
This commit also adds a Github workflow to check pull requests and adapts
copyright.pl to the changes.
Closes#8869
Also rephrase to make it sound less dangerous:
"ALPN: server did not agree on a protocol. Uses default."
Reported-by: Nick Coghlan
Fixes#8643Closes#8651
The TLS backends convert the host name to SNI name and need to use that.
This involves cutting off any trailing dot and lowercasing.
Co-authored-by: Jay Satiro
Closes#8320
Add support for `CURLOPT_CAINFO_BLOB` `CURLOPT_PROXY_CAINFO_BLOB` to the
rustls TLS backend. Multiple certificates in a single PEM string are
supported just like OpenSSL does with this option.
This is compatible at least with rustls-ffi 0.8+ which is our new
minimum version anyway.
I was able to build and run this on Windows, pulling trusted certs from
the system and then add them to rustls by setting
`CURLOPT_CAINFO_BLOB`. Handy!
Closes#8255
The comment is incorrect in two ways:
- It says the check needs to be last, but the check is actually first.
- is_handshaking actually starts out true.
Closes#8038
When we're reading out plaintext from rustls' internal buffers, we might
get a read of zero bytes (meaning a clean TCP close, including
close_notify). However, we shouldn't return immediately when that
happens, since we may have already copied out some plaintext bytes.
Break out of the loop when we get a read of zero bytes, and figure out
which path we're dealing with.
Acked-by: Kevin Burke
Closes#8003
The update to rustls-ffi 0.8.0 changed handling of EOF and close_notify.
From the CHANGELOG:
> Handling of unclean close and the close_notify TLS alert. Mirroring
> upstream changes, a rustls_connection now tracks TCP closed state like
> so: rustls_connection_read_tls considers a 0-length read from its
> callback to mean "TCP stream was closed by peer." If that happens
> before the peer sent close_notify, rustls_connection_read will return
> RUSTLS_RESULT_UNEXPECTED_EOF once the available plaintext bytes are
> exhausted. This is useful to protect against truncation attacks. Note:
> some TLS implementations don't send close_notify. If you are already
> getting length information from your protocol (e.g. Content-Length in
> HTTP) you may choose to ignore UNEXPECTED_EOF so long as the number of
> plaintext bytes was as expected.
That means we don't need to check for unclean EOF in `cr_recv()`,
because `process_new_packets()` will give us an error if appropriate.
Closes#8003
Previously we'd return CURLE_READ_ERROR if we received this, instead
of triggering the error handling logic that's present in the next if
block down.
After this change, curl requests to https://go.googlesource.com using
HTTP/2 complete successfully.
Fixes#7949Closes#7948
Some method names, as well as the generated library name, were changed
in a recent refactoring.
Further, change the default configuration instructions to check for
Hyper in either "target/debug" or "target/release" - the latter
contains an optimized build configuration.
Fixes#7947Closes#7948
We have and provide Curl_strerror() internally for a reason: strerror()
is not necessarily thread-safe so we should always try to avoid it.
Extended checksrc to warn for this, but feature the check disabled by
default and only enable it in lib/
Closes#7685
- the data needs to be "line-based" anyway since it's also passed to the
debug callback/application
- it makes infof() work like failf() and consistency is good
- there's an assert that triggers on newlines in the format string
- Also removes a few instances of "..."
- Removes the code that would append "..." to the end of the data *iff*
it was truncated in infof()
Closes#7357
For the commandline tool, we expect to be passed
SSL_CONN_CONFIG(CAfile); for library use, the use should pass a set of
trusted roots (like in other TLS backends).
This also removes a dependency on Security.framework when building on
macOS.
Closes#7250
The code in cr_recv was returning prematurely as soon as the socket
reported no more data to read. However, this could be leaving some
unread plaintext data in the rustls session from a previous call,
causing causing the transfer to hang if the socket never receives
further data.
We need to ensure that the session is fully drained of plaintext data
before returning CURLE_AGAIN to the caller.
Reviewed-by: Jacob Hoffman-Andrews
Closes#6894
If we get a close_notify, treat that as EOF. If we get an EOF from the
TCP stream, treat that as an error (because we should have ended the
connection earlier, when we got a close_notify).
Closes#6763
Previously, rustls was using an on-stack array for TLS data. However,
crustls has an (unusual) requirement that buffers it deals with are
initialized before writing to them. By using calloc, we can ensure the
buffer is initialized once and then reuse it across calls.
Closes#6742
This requires the latest main branch of crustls, which provides
rustls_client_config_builder_dangerous_set_certificate_verifier and
rustls_client_config_builder_set_enable_sni.
This refactors the session setup into its own function, and adds a new
function cr_hostname_is_ip. Because crustls doesn't support verification
of IP addresses, special handling is needed: We disable SNI and set a
placeholder hostname (which never actually gets sent on the wire).
Closes#6719
The Curl_easy pointer struct entry in connectdata is now gone. Just
before commit 215db086e0 landed on January 8, 2021 there were 919
references to conn->data.
Closes#6608
