Makes `Connection: keep-alive` behave correctly when making client
connections to UNIX domain sockets.
Prior to this, connections would never be re-used, but the keep-alive
would cause the connections to stick around until they time out. This
would lead to an eventual EMFILE error due to all the connections
staying open. This was due to http.Agent not properly supporting UNIX
domain sockets.
PR-URL: https://github.com/nodejs/node/pull/13214
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
The event manifest specifies the MethodID field as a 32 bit integer.
The 32 bit node executable publishes this correctly, but the 64 bit
executable published a 64 bit integer, making the event undecodable.
PR-URL: https://github.com/nodejs/node/pull/15563
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Hitesh Kanwathirtha <hiteshk@microsoft.com>
http2stream.respondWithFile api has changed since 8.5.0 with the
addition of the onError option. In the first code example an
onError function is implemented but never used, fix this
mistake.
Add a description to have more informations when onError is triggered.
PR-URL: https://github.com/nodejs/node/pull/15501
Fixes: https://github.com/nodejs/node/issues/15390
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Yuta Hiroto <hello@about-hiroppy.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
For consistency with the rest of the crypto classes, exposes the ECDH
class. Originally, only the createECDH function was exposed, and there
was no real reason to hide the class.
PR-URL: https://github.com/nodejs/node/pull/8188
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
The crypto classes are also exposed as createClass for each class. This
tests that each of them returns an instance of the class in question.
PR-URL: https://github.com/nodejs/node/pull/8188
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/15626
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
PR-URL: https://github.com/nodejs/node/pull/15592
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Daniel Bevenius <daniel.bevenius@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Sam Roberts <vieuxtech@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/15593
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Allow the user to pass in an execution_async_id instead of always
generating one. This way the JS API can be used to pre-allocate the
execution_async_id when the JS object is instantiated, before the native
resource is created.
Also allow the new execution_async_id to be passed via asyncReset().
PR-URL: https://github.com/nodejs/node/pull/14208
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Another optional argument is about to be added to AsyncWrap. So instead
of piling them on, create a separate constructor specifically for
PromiseWrap since it's the only class that uses the "silent" argument.
PR-URL: https://github.com/nodejs/node/pull/14208
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
The test addons/async-hooks-promise depends on there being only one hook
available. So skip it if NODE_TEST_WITH_ASYNC_HOOKS is set.
PR-URL: https://github.com/nodejs/node/pull/14208
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
When running tests with NODE_TEST_WITH_ASYNC_HOOKS and the same asyncId
is detected twice print the stack traces of both init() calls. Also
print if the resource is the same instance.
PR-URL: https://github.com/nodejs/node/pull/14208
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
The new test-with-async-hooks runs all normal tests (except async-hooks)
with the environment variable NODE_TEST_WITH_ASYNC_HOOKS set. These
extra checks do a minimum check to make sure async_hooks operates
normally under all other tests. e.g. if init() or destroy() is called
twice for the same id.
Also move test "async-hooks" from CI_JS_SUITES into its own
CI_ASYNC_HOOKS. Makes it cleaner to add, instead of supplying a massive
list of tests that may change in the future.
PR-URL: https://github.com/nodejs/node/pull/14208
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
If bootstrap throws and if ids are added to the async id stack and if
the exception wasn't handled by the fatal exception handler then the
AsyncCallbackScope destructor will cause the AsyncHooks::pop_ids() stack
check to fail. Causing the application to crash. So clear the async id
stack manually.
This is only possible if the user: 1) manually calls MakeCallback() or
2) uses async await in the top level. Which will cause _tickCallback()
to fire before bootstrap finishes executing.
The following example shows how the application can fail due to
exceeding the maximum call stack while using async await:
async function fn() {
fn();
throw new Error();
}
(async function() { await fn(); })();
If this occurs during bootstrap then the application will pring the
following warning a number of times then exit with a status of 0:
(node:*) UnhandledPromiseRejectionWarning: Unhandled promise rejection (rejection id: *): Error
Here's the same recursive call done after enabling a new AsyncHook() the
following will print instead of the above warning and exit with a
non-zero code (currently it's 7 because of how node::FatalException
assigns error codes based on where the failure happened):
script.js:25
async function fn() {
^
RangeError: Maximum call stack size exceeded
at <anonymous>
at fn (script.js:25:18)
at fn (script.js:26:3)
....
This has to do with how Promises lazily enable PromiseHook if an
AsyncHook() is enabled. Whether these need to be made uniform is outside
the scope of this commit
Fixes: https://github.com/nodejs/node/issues/15448
PR-URL: https://github.com/nodejs/node/pull/15553
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
The headers node_trace_writer.h and node_trace_buffer.h are not used in
agent.h but are more of an implementation detail of agent.cc.
This commit suggests moving the inclusion of these headers to agent.cc.
PR-URL: https://github.com/nodejs/node/pull/15598
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
The `countdown` and `fixtures` modules are documented in the
`common/README.md` file but are omitted from the table of contents. Add
them.
PR-URL: https://github.com/nodejs/node/pull/15595
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Turning FIPS mode on (or off) when it's already on (or off) should be a
no-op, not an error.
PR-URL: https://github.com/nodejs/node/pull/12210
Fixes: https://github.com/nodejs/node/issues/11849
Reviewed-By: Richard Lau <riclau@uk.ibm.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Fixes: https://github.com/nodejs/security/issues/147
PR-URL: https://github.com/nodejs-private/node-private/pull/94
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Doug Wilson <doug@somethingdoug.com>
Reviewed-By: Myles Borins <myles.borins@gmail.com>
Reviewed-By: Evan Lucas <evanlucas@me.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
code samples of napi_create_object and napi_property_descriptor were
not updated to latest API.
PR-URL: https://github.com/nodejs/node/pull/15581
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Handle edge case where stream pause is called between resume being
called and actually evaluated. Other minor adjustments to avoid
various edge cases around stream events. Add new tests that cover
all changes.
Fixes: https://github.com/nodejs/node/issues/15491
PR-URL: https://github.com/nodejs/node/pull/15503
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Sections in the https://tools.ietf.org/html/rfc1122 are not linkable
anymore. Replace the possibly confusing hash with the explicit
section number in the text.
PR-URL: https://github.com/nodejs/node/pull/15542
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Address issue with Windows drive letter handling that was
causing es-module test suite to fail.
PR-URL: https://github.com/nodejs/node/pull/15490
Ref: https://github.com/whatwg/url/pull/343
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Expand existing test cases to hit some additional branches of
toHeaderObject and mapToHeaders.
PR-URL: https://github.com/nodejs/node/pull/15493
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
We copy the "a.js" file fixture to a temp directory and then we open
it with write access. This makes sure we are only writing to the
provided temp-dir and not in the fixture dir.
PR-URL: https://github.com/nodejs/node/pull/15537
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
AliasedBuffer tests weren't creating an v8::Isolate::Scope, and this
had negative impact on the node-chakracore branch, where expectation
is an Isoalte has been Enter()'d before being used.
PR-URL: https://github.com/nodejs/node/pull/15536
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Replace a somewhat idiosyncratic use of `split()` to remove a prefix
with `replace()`. (A case could be made for `slice()` as well but I
think this is more readable.)
PR-URL: https://github.com/nodejs/node/pull/15545
Reviewed-By: Jeremiah Senkpiel <fishrock123@rocketmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
`invalidArgType()` uses `includes()` in two places where `startsWith()`
and `endsWith()` are more appropriate (at least in my opinion). Switch
to those more specific functions.
PR-URL: https://github.com/nodejs/node/pull/15544
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Minwoo Jung <minwoo@nodesource.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Remove unused onTimeout on Http2Session and Http2Stream because
the correct _onTimeout is already declared and in use. Expand
timeout tests to handle edge cases and additional arguments.
PR-URL: https://github.com/nodejs/node/pull/15539
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
"jslint" is the name of a tool that actually is not used, which can
cause confusion.
PR-URL: https://github.com/nodejs/node/pull/15272
Reviewed-By: Roman Reiss <me@silverwind.io>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Gibson Fahnestock <gibfahn@gmail.com>
Reviewed-By: Refael Ackermann <refack@gmail.com>
Reviewed-By: Alexey Orlenko <eaglexrlnk@gmail.com>
Define BUILDING_NGHTTP2 in order that NGHTTP2_EXTERN is properly defined
when building the nghttp2 static library.
Move NGHTTP2_STATICLIB out of node.gyp because it is a property of the
nghttp2 static library, not the node executable.
PR-URL: https://github.com/nodejs/node/pull/15487
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/15508
Reviewed-By: Michael Dawson <michael_dawson@ca.ibm.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Franziska Hinkelmann <franziska.hinkelmann@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
This change fixes a potential segfault in the sampling heap profiler.
This landed as part of a larger change upstream [1]. This is the minimal
backport that avoids the segfault.
[1]: https://git.io/vdTYL
PR-URL: https://github.com/nodejs/node/pull/15498
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
- Move `module_map` to `Environment` instead of having it be global
state
- `std::map` → `std::unordered_map`
- Remove one level of indirection for the map values
- Clean up empty vectors in `module_map`
- Call `Reset()` on all persistent handles in `resolve_cache_`
- Add a missing `HandleScope` to `ModuleWrap::~ModuleWrap()`
PR-URL: https://github.com/nodejs/node/pull/15515
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Bradley Farias <bradley.meck@gmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Add a `promiseResolve()` hook.
PR-URL: https://github.com/nodejs/node/pull/15296
Reviewed-By: Trevor Norris <trev.norris@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Adds the following tests for tty readline:
- go to beginning and end of line
- wordLeft
- wordRight
- deleteWordLeft
- deleteWordRight
PR-URL: https://github.com/nodejs/node/pull/12064
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
The _debug of Module is undocumented and it useless here.
PR-URL: https://github.com/nodejs/node/pull/13948
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Timothy Gu <timothygu99@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Makes tls.connect() behave as documented, preferring options.path over
options.port. This makes it consistent with net.connect(), so the
included test demonstrates that both behave in this way.
Also, for consistency, noting the precedence of options.path in net
doc.
PR-URL: https://github.com/nodejs/node/pull/14564
Reviewed-By: Sakthipriyan Vairamani <thechargingvolcano@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
The message sent using process.send() goes through JSON
serialization and parsing, which could lead to surprising behaviors.
This commit elaborate a bit more on this and add a link to
the notes about these caveats in the ECMAScript specification.
PR-URL: https://github.com/nodejs/node/pull/12963
Refs: https://github.com/nodejs/node/issues/12497
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Vse Mozhet Byt <vsemozhetbyt@gmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>