curl/tests/http/testenv/env.py

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

585 lines
18 KiB
Python
Raw Normal View History

#!/usr/bin/env python3
# -*- coding: utf-8 -*-
#***************************************************************************
# _ _ ____ _
# Project ___| | | | _ \| |
# / __| | | | |_) | |
# | (__| |_| | _ <| |___
# \___|\___/|_| \_\_____|
#
# Copyright (C) Daniel Stenberg, <daniel@haxx.se>, et al.
#
# This software is licensed as described in the file COPYING, which
# you should have received as part of this distribution. The terms
# are also available at https://curl.se/docs/copyright.html.
#
# You may opt to use, copy, modify, merge, publish, distribute and/or sell
# copies of the Software, and permit persons to whom the Software is
# furnished to do so, under the terms of the COPYING file.
#
# This software is distributed on an "AS IS" basis, WITHOUT WARRANTY OF ANY
# KIND, either express or implied.
#
# SPDX-License-Identifier: curl
#
###########################################################################
#
import logging
import os
import re
lib: graceful connection shutdown When libcurl discards a connection there are two phases this may go through: "shutdown" and "closing". If a connection is aborted, the shutdown phase is skipped and it is closed right away. The connection filters attached to the connection implement the phases in their `do_shutdown()` and `do_close()` callbacks. Filters carry now a `shutdown` flags next to `connected` to keep track of the shutdown operation. Filters are shut down from top to bottom. If a filter is not connected, its shutdown is skipped. Notable filters that *do* something during shutdown are HTTP/2 and TLS. HTTP/2 sends the GOAWAY frame. TLS sends its close notify and expects to receive a close notify from the server. As sends and receives may EAGAIN on the network, a shutdown is often not successful right away and needs to poll the connection's socket(s). To facilitate this, such connections are placed on a new shutdown list inside the connection cache. Since managing this list requires the cooperation of a multi handle, only the connection cache belonging to a multi handle is used. If a connection was in another cache when being discarded, it is removed there and added to the multi's cache. If no multi handle is available at that time, the connection is shutdown and closed in a one-time, best-effort attempt. When a multi handle is destroyed, all connection still on the shutdown list are discarded with a final shutdown attempt and close. In curl debug builds, the environment variable `CURL_GRACEFUL_SHUTDOWN` can be set to make this graceful with a timeout in milliseconds given by the variable. The shutdown list is limited to the max number of connections configured for a multi cache. Set via CURLMOPT_MAX_TOTAL_CONNECTIONS. When the limit is reached, the oldest connection on the shutdown list is discarded. - In multi_wait() and multi_waitfds(), collect all connection caches involved (each transfer might carry its own) into a temporary list. Let each connection cache on the list contribute sockets and POLLIN/OUT events it's connections are waiting for. - in multi_perform() collect the connection caches the same way and let them peform their maintenance. This will make another non-blocking attempt to shutdown all connections on its shutdown list. - for event based multis (multi->socket_cb set), add the sockets and their poll events via the callback. When `multi_socket()` is invoked for a socket not known by an active transfer, forward this to the multi's cache for processing. On closing a connection, remove its socket(s) via the callback. TLS connection filters MUST NOT send close nofity messages in their `do_close()` implementation. The reason is that a TLS close notify signals a success. When a connection is aborted and skips its shutdown phase, the server needs to see a missing close notify to detect something has gone wrong. A graceful shutdown of FTP's data connection is performed implicitly before regarding the upload/download as complete and continuing on the control connection. For FTP without TLS, there is just the socket close happening. But with TLS, the sent/received close notify signals that the transfer is complete and healthy. Servers like `vsftpd` verify that and reject uploads without a TLS close notify. - added test_19_* for shutdown related tests - test_19_01 and test_19_02 test for TCP RST packets which happen without a graceful shutdown and should no longer appear otherwise. - add test_19_03 for handling shutdowns by the server - add test_19_04 for handling shutdowns by curl - add test_19_05 for event based shutdowny by server - add test_30_06/07 and test_31_06/07 for shutdown checks on FTP up- and downloads. Closes #13976
2024-06-19 18:40:06 +08:00
import shutil
import socket
import subprocess
import sys
from configparser import ConfigParser, ExtendedInterpolation
from datetime import timedelta
from typing import Optional
import pytest
from .certs import CertificateSpec, TestCA, Credentials
from .ports import alloc_ports
log = logging.getLogger(__name__)
def init_config_from(conf_path):
if os.path.isfile(conf_path):
config = ConfigParser(interpolation=ExtendedInterpolation())
config.read(conf_path)
return config
return None
TESTS_HTTPD_PATH = os.path.dirname(os.path.dirname(__file__))
DEF_CONFIG = init_config_from(os.path.join(TESTS_HTTPD_PATH, 'config.ini'))
TOP_PATH = os.path.dirname(os.path.dirname(TESTS_HTTPD_PATH))
CURL = os.path.join(TOP_PATH, 'src/curl')
class EnvConfig:
def __init__(self):
self.tests_dir = TESTS_HTTPD_PATH
self.gen_dir = os.path.join(self.tests_dir, 'gen')
self.project_dir = os.path.dirname(os.path.dirname(self.tests_dir))
self.config = DEF_CONFIG
# check cur and its features
self.curl = CURL
if 'CURL' in os.environ:
self.curl = os.environ['CURL']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
self.curl_props = {
'version': None,
'os': None,
'fullname': None,
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
'features': [],
'protocols': [],
'libs': [],
'lib_versions': [],
}
self.curl_is_debug = False
self.curl_protos = []
p = subprocess.run(args=[self.curl, '-V'],
capture_output=True, text=True)
if p.returncode != 0:
assert False, f'{self.curl} -V failed with exit code: {p.returncode}'
if p.stderr.startswith('WARNING:'):
self.curl_is_debug = True
for l in p.stdout.splitlines(keepends=False):
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
if l.startswith('curl '):
m = re.match(r'^curl (?P<version>\S+) (?P<os>\S+) (?P<libs>.*)$', l)
if m:
self.curl_props['fullname'] = m.group(0)
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
self.curl_props['version'] = m.group('version')
self.curl_props['os'] = m.group('os')
self.curl_props['lib_versions'] = [
lib.lower() for lib in m.group('libs').split(' ')
]
self.curl_props['libs'] = [
re.sub(r'/.*', '', lib) for lib in self.curl_props['lib_versions']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
]
if l.startswith('Features: '):
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
self.curl_props['features'] = [
feat.lower() for feat in l[10:].split(' ')
]
if l.startswith('Protocols: '):
self.curl_props['protocols'] = [
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
prot.lower() for prot in l[11:].split(' ')
]
self.ports = alloc_ports(port_specs={
'ftp': socket.SOCK_STREAM,
'ftps': socket.SOCK_STREAM,
'http': socket.SOCK_STREAM,
'https': socket.SOCK_STREAM,
'proxy': socket.SOCK_STREAM,
'proxys': socket.SOCK_STREAM,
'h2proxys': socket.SOCK_STREAM,
'caddy': socket.SOCK_STREAM,
'caddys': socket.SOCK_STREAM,
'ws': socket.SOCK_STREAM,
})
self.httpd = self.config['httpd']['httpd']
self.apachectl = self.config['httpd']['apachectl']
self.apxs = self.config['httpd']['apxs']
if len(self.apxs) == 0:
self.apxs = None
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
self._httpd_version = None
self.examples_pem = {
'key': 'xxx',
'cert': 'xxx',
}
self.htdocs_dir = os.path.join(self.gen_dir, 'htdocs')
self.tld = 'http.curl.se'
self.domain1 = f"one.{self.tld}"
self.domain1brotli = f"brotli.one.{self.tld}"
self.domain2 = f"two.{self.tld}"
self.ftp_domain = f"ftp.{self.tld}"
self.proxy_domain = f"proxy.{self.tld}"
self.cert_specs = [
CertificateSpec(domains=[self.domain1, self.domain1brotli, 'localhost', '127.0.0.1'], key_type='rsa2048'),
CertificateSpec(domains=[self.domain2], key_type='rsa2048'),
CertificateSpec(domains=[self.ftp_domain], key_type='rsa2048'),
CertificateSpec(domains=[self.proxy_domain, '127.0.0.1'], key_type='rsa2048'),
CertificateSpec(name="clientsX", sub_specs=[
CertificateSpec(name="user1", client=True),
]),
]
self.nghttpx = self.config['nghttpx']['nghttpx']
if len(self.nghttpx.strip()) == 0:
self.nghttpx = None
self._nghttpx_version = None
self.nghttpx_with_h3 = False
if self.nghttpx is not None:
p = subprocess.run(args=[self.nghttpx, '-v'],
capture_output=True, text=True)
if p.returncode != 0:
# not a working nghttpx
self.nghttpx = None
else:
self._nghttpx_version = re.sub(r'^nghttpx\s*', '', p.stdout.strip())
self.nghttpx_with_h3 = re.match(r'.* nghttp3/.*', p.stdout.strip()) is not None
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
log.debug(f'nghttpx -v: {p.stdout}')
self.caddy = self.config['caddy']['caddy']
self._caddy_version = None
if len(self.caddy.strip()) == 0:
self.caddy = None
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
if self.caddy is not None:
try:
p = subprocess.run(args=[self.caddy, 'version'],
capture_output=True, text=True)
if p.returncode != 0:
# not a working caddy
self.caddy = None
m = re.match(r'v?(\d+\.\d+\.\d+).*', p.stdout)
if m:
self._caddy_version = m.group(1)
else:
raise f'Unable to determine cadd version from: {p.stdout}'
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
except:
self.caddy = None
self.vsftpd = self.config['vsftpd']['vsftpd']
self._vsftpd_version = None
if self.vsftpd is not None:
try:
p = subprocess.run(args=[self.vsftpd, '-v'],
capture_output=True, text=True)
if p.returncode != 0:
# not a working vsftpd
self.vsftpd = None
m = re.match(r'vsftpd: version (\d+\.\d+\.\d+)', p.stderr)
if m:
self._vsftpd_version = m.group(1)
elif len(p.stderr) == 0:
# vsftp does not use stdout or stderr for printing its version... -.-
self._vsftpd_version = 'unknown'
else:
raise Exception(f'Unable to determine VsFTPD version from: {p.stderr}')
except Exception as e:
self.vsftpd = None
lib: graceful connection shutdown When libcurl discards a connection there are two phases this may go through: "shutdown" and "closing". If a connection is aborted, the shutdown phase is skipped and it is closed right away. The connection filters attached to the connection implement the phases in their `do_shutdown()` and `do_close()` callbacks. Filters carry now a `shutdown` flags next to `connected` to keep track of the shutdown operation. Filters are shut down from top to bottom. If a filter is not connected, its shutdown is skipped. Notable filters that *do* something during shutdown are HTTP/2 and TLS. HTTP/2 sends the GOAWAY frame. TLS sends its close notify and expects to receive a close notify from the server. As sends and receives may EAGAIN on the network, a shutdown is often not successful right away and needs to poll the connection's socket(s). To facilitate this, such connections are placed on a new shutdown list inside the connection cache. Since managing this list requires the cooperation of a multi handle, only the connection cache belonging to a multi handle is used. If a connection was in another cache when being discarded, it is removed there and added to the multi's cache. If no multi handle is available at that time, the connection is shutdown and closed in a one-time, best-effort attempt. When a multi handle is destroyed, all connection still on the shutdown list are discarded with a final shutdown attempt and close. In curl debug builds, the environment variable `CURL_GRACEFUL_SHUTDOWN` can be set to make this graceful with a timeout in milliseconds given by the variable. The shutdown list is limited to the max number of connections configured for a multi cache. Set via CURLMOPT_MAX_TOTAL_CONNECTIONS. When the limit is reached, the oldest connection on the shutdown list is discarded. - In multi_wait() and multi_waitfds(), collect all connection caches involved (each transfer might carry its own) into a temporary list. Let each connection cache on the list contribute sockets and POLLIN/OUT events it's connections are waiting for. - in multi_perform() collect the connection caches the same way and let them peform their maintenance. This will make another non-blocking attempt to shutdown all connections on its shutdown list. - for event based multis (multi->socket_cb set), add the sockets and their poll events via the callback. When `multi_socket()` is invoked for a socket not known by an active transfer, forward this to the multi's cache for processing. On closing a connection, remove its socket(s) via the callback. TLS connection filters MUST NOT send close nofity messages in their `do_close()` implementation. The reason is that a TLS close notify signals a success. When a connection is aborted and skips its shutdown phase, the server needs to see a missing close notify to detect something has gone wrong. A graceful shutdown of FTP's data connection is performed implicitly before regarding the upload/download as complete and continuing on the control connection. For FTP without TLS, there is just the socket close happening. But with TLS, the sent/received close notify signals that the transfer is complete and healthy. Servers like `vsftpd` verify that and reject uploads without a TLS close notify. - added test_19_* for shutdown related tests - test_19_01 and test_19_02 test for TCP RST packets which happen without a graceful shutdown and should no longer appear otherwise. - add test_19_03 for handling shutdowns by the server - add test_19_04 for handling shutdowns by curl - add test_19_05 for event based shutdowny by server - add test_30_06/07 and test_31_06/07 for shutdown checks on FTP up- and downloads. Closes #13976
2024-06-19 18:40:06 +08:00
self._tcpdump = shutil.which('tcpdump')
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@property
def httpd_version(self):
if self._httpd_version is None and self.apxs is not None:
try:
p = subprocess.run(args=[self.apxs, '-q', 'HTTPD_VERSION'],
capture_output=True, text=True)
if p.returncode != 0:
log.error(f'{self.apxs} failed to query HTTPD_VERSION: {p}')
else:
self._httpd_version = p.stdout.strip()
except Exception as e:
log.error(f'{self.apxs} failed to run: {e}')
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
return self._httpd_version
def versiontuple(self, v):
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
v = re.sub(r'(\d+\.\d+(\.\d+)?)(-\S+)?', r'\1', v)
return tuple(map(int, v.split('.')))
def httpd_is_at_least(self, minv):
if self.httpd_version is None:
return False
hv = self.versiontuple(self.httpd_version)
return hv >= self.versiontuple(minv)
def caddy_is_at_least(self, minv):
if self.caddy_version is None:
return False
hv = self.versiontuple(self.caddy_version)
return hv >= self.versiontuple(minv)
def is_complete(self) -> bool:
return os.path.isfile(self.httpd) and \
os.path.isfile(self.apachectl) and \
self.apxs is not None and \
os.path.isfile(self.apxs)
def get_incomplete_reason(self) -> Optional[str]:
if self.httpd is None or len(self.httpd.strip()) == 0:
return f'httpd not configured, see `--with-test-httpd=<path>`'
if not os.path.isfile(self.httpd):
return f'httpd ({self.httpd}) not found'
if not os.path.isfile(self.apachectl):
return f'apachectl ({self.apachectl}) not found'
if self.apxs is None:
return f"command apxs not found (commonly provided in apache2-dev)"
if not os.path.isfile(self.apxs):
return f"apxs ({self.apxs}) not found"
return None
@property
def nghttpx_version(self):
return self._nghttpx_version
@property
def caddy_version(self):
return self._caddy_version
@property
def vsftpd_version(self):
return self._vsftpd_version
lib: graceful connection shutdown When libcurl discards a connection there are two phases this may go through: "shutdown" and "closing". If a connection is aborted, the shutdown phase is skipped and it is closed right away. The connection filters attached to the connection implement the phases in their `do_shutdown()` and `do_close()` callbacks. Filters carry now a `shutdown` flags next to `connected` to keep track of the shutdown operation. Filters are shut down from top to bottom. If a filter is not connected, its shutdown is skipped. Notable filters that *do* something during shutdown are HTTP/2 and TLS. HTTP/2 sends the GOAWAY frame. TLS sends its close notify and expects to receive a close notify from the server. As sends and receives may EAGAIN on the network, a shutdown is often not successful right away and needs to poll the connection's socket(s). To facilitate this, such connections are placed on a new shutdown list inside the connection cache. Since managing this list requires the cooperation of a multi handle, only the connection cache belonging to a multi handle is used. If a connection was in another cache when being discarded, it is removed there and added to the multi's cache. If no multi handle is available at that time, the connection is shutdown and closed in a one-time, best-effort attempt. When a multi handle is destroyed, all connection still on the shutdown list are discarded with a final shutdown attempt and close. In curl debug builds, the environment variable `CURL_GRACEFUL_SHUTDOWN` can be set to make this graceful with a timeout in milliseconds given by the variable. The shutdown list is limited to the max number of connections configured for a multi cache. Set via CURLMOPT_MAX_TOTAL_CONNECTIONS. When the limit is reached, the oldest connection on the shutdown list is discarded. - In multi_wait() and multi_waitfds(), collect all connection caches involved (each transfer might carry its own) into a temporary list. Let each connection cache on the list contribute sockets and POLLIN/OUT events it's connections are waiting for. - in multi_perform() collect the connection caches the same way and let them peform their maintenance. This will make another non-blocking attempt to shutdown all connections on its shutdown list. - for event based multis (multi->socket_cb set), add the sockets and their poll events via the callback. When `multi_socket()` is invoked for a socket not known by an active transfer, forward this to the multi's cache for processing. On closing a connection, remove its socket(s) via the callback. TLS connection filters MUST NOT send close nofity messages in their `do_close()` implementation. The reason is that a TLS close notify signals a success. When a connection is aborted and skips its shutdown phase, the server needs to see a missing close notify to detect something has gone wrong. A graceful shutdown of FTP's data connection is performed implicitly before regarding the upload/download as complete and continuing on the control connection. For FTP without TLS, there is just the socket close happening. But with TLS, the sent/received close notify signals that the transfer is complete and healthy. Servers like `vsftpd` verify that and reject uploads without a TLS close notify. - added test_19_* for shutdown related tests - test_19_01 and test_19_02 test for TCP RST packets which happen without a graceful shutdown and should no longer appear otherwise. - add test_19_03 for handling shutdowns by the server - add test_19_04 for handling shutdowns by curl - add test_19_05 for event based shutdowny by server - add test_30_06/07 and test_31_06/07 for shutdown checks on FTP up- and downloads. Closes #13976
2024-06-19 18:40:06 +08:00
@property
def tcpdmp(self) -> Optional[str]:
return self._tcpdump
class Env:
CONFIG = EnvConfig()
@staticmethod
def setup_incomplete() -> bool:
return not Env.CONFIG.is_complete()
@staticmethod
def incomplete_reason() -> Optional[str]:
return Env.CONFIG.get_incomplete_reason()
@staticmethod
def have_nghttpx() -> bool:
return Env.CONFIG.nghttpx is not None
@staticmethod
def have_h3_server() -> bool:
return Env.CONFIG.nghttpx_with_h3
@staticmethod
def have_ssl_curl() -> bool:
return 'ssl' in Env.CONFIG.curl_props['features']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def have_h2_curl() -> bool:
return 'http2' in Env.CONFIG.curl_props['features']
@staticmethod
def have_h3_curl() -> bool:
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
return 'http3' in Env.CONFIG.curl_props['features']
@staticmethod
def curl_uses_lib(libname: str) -> bool:
return libname.lower() in Env.CONFIG.curl_props['libs']
@staticmethod
def curl_uses_ossl_quic() -> bool:
if Env.have_h3_curl():
return not Env.curl_uses_lib('ngtcp2') and Env.curl_uses_lib('nghttp3')
return False
@staticmethod
def curl_has_feature(feature: str) -> bool:
return feature.lower() in Env.CONFIG.curl_props['features']
@staticmethod
def curl_has_protocol(protocol: str) -> bool:
return protocol.lower() in Env.CONFIG.curl_props['protocols']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def curl_lib_version(libname: str) -> str:
prefix = f'{libname.lower()}/'
for lversion in Env.CONFIG.curl_props['lib_versions']:
if lversion.startswith(prefix):
return lversion[len(prefix):]
return 'unknown'
@staticmethod
def curl_lib_version_at_least(libname: str, min_version) -> str:
lversion = Env.curl_lib_version(libname)
if lversion != 'unknown':
return Env.CONFIG.versiontuple(min_version) <= \
Env.CONFIG.versiontuple(lversion)
return False
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def curl_os() -> str:
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
return Env.CONFIG.curl_props['os']
@staticmethod
def curl_fullname() -> str:
return Env.CONFIG.curl_props['fullname']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def curl_version() -> str:
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
return Env.CONFIG.curl_props['version']
@staticmethod
def curl_is_debug() -> bool:
return Env.CONFIG.curl_is_debug
@staticmethod
def have_h3() -> bool:
return Env.have_h3_curl() and Env.have_h3_server()
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def httpd_version() -> str:
return Env.CONFIG.httpd_version
@staticmethod
def nghttpx_version() -> str:
return Env.CONFIG.nghttpx_version
@staticmethod
def caddy_version() -> str:
return Env.CONFIG.caddy_version
@staticmethod
def caddy_is_at_least(minv) -> bool:
return Env.CONFIG.caddy_is_at_least(minv)
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@staticmethod
def httpd_is_at_least(minv) -> bool:
return Env.CONFIG.httpd_is_at_least(minv)
@staticmethod
def has_caddy() -> bool:
return Env.CONFIG.caddy is not None
@staticmethod
def has_vsftpd() -> bool:
return Env.CONFIG.vsftpd is not None
@staticmethod
def vsftpd_version() -> str:
return Env.CONFIG.vsftpd_version
lib: graceful connection shutdown When libcurl discards a connection there are two phases this may go through: "shutdown" and "closing". If a connection is aborted, the shutdown phase is skipped and it is closed right away. The connection filters attached to the connection implement the phases in their `do_shutdown()` and `do_close()` callbacks. Filters carry now a `shutdown` flags next to `connected` to keep track of the shutdown operation. Filters are shut down from top to bottom. If a filter is not connected, its shutdown is skipped. Notable filters that *do* something during shutdown are HTTP/2 and TLS. HTTP/2 sends the GOAWAY frame. TLS sends its close notify and expects to receive a close notify from the server. As sends and receives may EAGAIN on the network, a shutdown is often not successful right away and needs to poll the connection's socket(s). To facilitate this, such connections are placed on a new shutdown list inside the connection cache. Since managing this list requires the cooperation of a multi handle, only the connection cache belonging to a multi handle is used. If a connection was in another cache when being discarded, it is removed there and added to the multi's cache. If no multi handle is available at that time, the connection is shutdown and closed in a one-time, best-effort attempt. When a multi handle is destroyed, all connection still on the shutdown list are discarded with a final shutdown attempt and close. In curl debug builds, the environment variable `CURL_GRACEFUL_SHUTDOWN` can be set to make this graceful with a timeout in milliseconds given by the variable. The shutdown list is limited to the max number of connections configured for a multi cache. Set via CURLMOPT_MAX_TOTAL_CONNECTIONS. When the limit is reached, the oldest connection on the shutdown list is discarded. - In multi_wait() and multi_waitfds(), collect all connection caches involved (each transfer might carry its own) into a temporary list. Let each connection cache on the list contribute sockets and POLLIN/OUT events it's connections are waiting for. - in multi_perform() collect the connection caches the same way and let them peform their maintenance. This will make another non-blocking attempt to shutdown all connections on its shutdown list. - for event based multis (multi->socket_cb set), add the sockets and their poll events via the callback. When `multi_socket()` is invoked for a socket not known by an active transfer, forward this to the multi's cache for processing. On closing a connection, remove its socket(s) via the callback. TLS connection filters MUST NOT send close nofity messages in their `do_close()` implementation. The reason is that a TLS close notify signals a success. When a connection is aborted and skips its shutdown phase, the server needs to see a missing close notify to detect something has gone wrong. A graceful shutdown of FTP's data connection is performed implicitly before regarding the upload/download as complete and continuing on the control connection. For FTP without TLS, there is just the socket close happening. But with TLS, the sent/received close notify signals that the transfer is complete and healthy. Servers like `vsftpd` verify that and reject uploads without a TLS close notify. - added test_19_* for shutdown related tests - test_19_01 and test_19_02 test for TCP RST packets which happen without a graceful shutdown and should no longer appear otherwise. - add test_19_03 for handling shutdowns by the server - add test_19_04 for handling shutdowns by curl - add test_19_05 for event based shutdowny by server - add test_30_06/07 and test_31_06/07 for shutdown checks on FTP up- and downloads. Closes #13976
2024-06-19 18:40:06 +08:00
@staticmethod
def tcpdump() -> Optional[str]:
return Env.CONFIG.tcpdmp
def __init__(self, pytestconfig=None):
self._verbose = pytestconfig.option.verbose \
if pytestconfig is not None else 0
self._ca = None
self._test_timeout = 300.0 if self._verbose > 1 else 60.0 # seconds
def issue_certs(self):
if self._ca is None:
ca_dir = os.path.join(self.CONFIG.gen_dir, 'ca')
self._ca = TestCA.create_root(name=self.CONFIG.tld,
store_dir=ca_dir,
key_type="rsa2048")
self._ca.issue_certs(self.CONFIG.cert_specs)
def setup(self):
os.makedirs(self.gen_dir, exist_ok=True)
os.makedirs(self.htdocs_dir, exist_ok=True)
self.issue_certs()
def get_credentials(self, domain) -> Optional[Credentials]:
creds = self.ca.get_credentials_for_name(domain)
if len(creds) > 0:
return creds[0]
return None
@property
def verbose(self) -> int:
return self._verbose
@property
def test_timeout(self) -> Optional[float]:
return self._test_timeout
@test_timeout.setter
def test_timeout(self, val: Optional[float]):
self._test_timeout = val
@property
def gen_dir(self) -> str:
return self.CONFIG.gen_dir
@property
def project_dir(self) -> str:
return self.CONFIG.project_dir
@property
def ca(self):
return self._ca
@property
def htdocs_dir(self) -> str:
return self.CONFIG.htdocs_dir
@property
def domain1(self) -> str:
return self.CONFIG.domain1
@property
def domain1brotli(self) -> str:
return self.CONFIG.domain1brotli
@property
def domain2(self) -> str:
return self.CONFIG.domain2
@property
def ftp_domain(self) -> str:
return self.CONFIG.ftp_domain
@property
def proxy_domain(self) -> str:
return self.CONFIG.proxy_domain
@property
def http_port(self) -> int:
return self.CONFIG.ports['http']
@property
def https_port(self) -> int:
return self.CONFIG.ports['https']
@property
def h3_port(self) -> int:
return self.https_port
@property
def proxy_port(self) -> int:
return self.CONFIG.ports['proxy']
@property
def proxys_port(self) -> int:
return self.CONFIG.ports['proxys']
@property
def ftp_port(self) -> int:
return self.CONFIG.ports['ftp']
@property
def ftps_port(self) -> int:
return self.CONFIG.ports['ftps']
@property
def h2proxys_port(self) -> int:
return self.CONFIG.ports['h2proxys']
def pts_port(self, proto: str = 'http/1.1') -> int:
# proxy tunnel port
return self.CONFIG.ports['h2proxys' if proto == 'h2' else 'proxys']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@property
def caddy(self) -> str:
return self.CONFIG.caddy
@property
def caddy_https_port(self) -> int:
return self.CONFIG.ports['caddys']
@property
def caddy_http_port(self) -> int:
return self.CONFIG.ports['caddy']
connections: introduce http/3 happy eyeballs 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
2023-02-02 00:13:12 +08:00
@property
def vsftpd(self) -> str:
return self.CONFIG.vsftpd
@property
def ws_port(self) -> int:
return self.CONFIG.ports['ws']
@property
def curl(self) -> str:
return self.CONFIG.curl
@property
def httpd(self) -> str:
return self.CONFIG.httpd
@property
def apachectl(self) -> str:
return self.CONFIG.apachectl
@property
def apxs(self) -> str:
return self.CONFIG.apxs
@property
def nghttpx(self) -> Optional[str]:
return self.CONFIG.nghttpx
@property
def slow_network(self) -> bool:
return "CURL_DBG_SOCK_WBLOCK" in os.environ or \
"CURL_DBG_SOCK_WPARTIAL" in os.environ
@property
def ci_run(self) -> bool:
return "CURL_CI" in os.environ
def authority_for(self, domain: str, alpn_proto: Optional[str] = None):
if alpn_proto is None or \
alpn_proto in ['h2', 'http/1.1', 'http/1.0', 'http/0.9']:
return f'{domain}:{self.https_port}'
if alpn_proto in ['h3']:
return f'{domain}:{self.h3_port}'
return f'{domain}:{self.http_port}'
def make_data_file(self, indir: str, fname: str, fsize: int) -> str:
fpath = os.path.join(indir, fname)
s10 = "0123456789"
s = (101 * s10) + s10[0:3]
with open(fpath, 'w') as fd:
for i in range(int(fsize / 1024)):
fd.write(f"{i:09d}-{s}\n")
remain = int(fsize % 1024)
if remain != 0:
i = int(fsize / 1024) + 1
s = f"{i:09d}-{s}\n"
fd.write(s[0:remain])
return fpath
def make_clients(self):
client_dir = os.path.join(self.project_dir, 'tests/http/clients')
p = subprocess.run(['make'], capture_output=True, text=True,
cwd=client_dir)
if p.returncode != 0:
pytest.exit(f"`make`in {client_dir} failed:\n{p.stderr}")
return False
return True