`http.request({ highWaterMark })` passes the value to the TCP socket
via createConnection() but does not set it on the OutgoingMessage
internal kHighWaterMark. OutgoingMessage._writeRaw() has two mutually
exclusive write paths:
Path A (socket connected): conn.write() — uses socket HWM ✓
Path B (no socket yet): outputSize < this[kHighWaterMark] — uses
OutgoingMessage own default (64 KB) ✗
Because the OutgoingMessage constructor already accepts
options.highWaterMark, the fix is to set kHighWaterMark from the
user options after they are parsed in the ClientRequest constructor.
This resolves two symptoms:
1. write() returning the wrong boolean for pre-socket writes (the
user highWaterMark was silently ignored on all Node versions).
2. A deadlock on Node >= 24.16.0 where the incorrect false return
sets kNeedDrain, but drain never fires because the socket was
never backpressured (introduced by the stricter drain gate in
#62936).
Signed-off-by: Naman Trivedi <trivenay@amazon.com>
Fixes: https://github.com/nodejs/node/issues/64645
Refs: https://github.com/nodejs/node/pull/62936
PR-URL: https://github.com/nodejs/node/pull/64653
Reviewed-By: Trivikram Kamat <trivikr.dev@gmail.com>
Reviewed-By: Tim Perry <pimterry@gmail.com>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
When sqlite3_exec() or sqlite3changeset_apply() call JavaScript
callbacks (user-defined functions, conflict handlers, or filter
callbacks), the DatabaseSync object could be garbage-collected
if the JavaScript code drops all references to it. Both methods
only held a raw DatabaseSync* pointer on the C++ stack, which
V8 GC does not track.
Add a BaseObjectPtr<DatabaseSync> guard that keeps the database
alive for the duration of these SQLite API calls, preventing a
use-after-free when the JavaScript callback triggers GC.
Signed-off-by: Matteo Collina <hello@matteocollina.com>
PR-URL: https://github.com/nodejs/node/pull/64535
Reviewed-By: Edy Silva <edigleyssonsilva@gmail.com>
Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: René <contact.9a5d6388@renegade334.me.uk>
This ensures we don't fire session events for totally invalid TLS
handshakes - fundamental errors, bad SNI/ALPN values, or anything else
that our TLS config would reject. Instead, it means servers can access
servername & alpnProtocol synchronously as soon as the event is fired -
all key session data is available and it's immediately usable.
We don't want to defer further to handshake completed, since that'd be
an extra RT, and defeat 0RTT benefits entirely. ClientHello processed
without errors is sufficient for now.
This isn't a security mechanism. Existing structures will defer
actually sending & receiving anything that's not marked explicitly as
early data until the handshake completes anyway.
Signed-off-by: Tim Perry <pimterry@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/64132
Reviewed-By: James M Snell <jasnell@gmail.com>
Every string-named JavaScript channel consumed an entry in the fixed
native subscriber array. Creating more than 1,024 channels triggered a
CHECK and terminated the process.
Allocate slots only for native publishers and grow the aliased buffer
when it fills. Refresh the JavaScript view after resizing and preserve
the capacity in snapshots.
Signed-off-by: Stephen Belanger <admin@stephenbelanger.com>
PR-URL: https://github.com/nodejs/node/pull/64497
Reviewed-By: Rafael Gonzaga <rafael.nunu@hotmail.com>
Reviewed-By: Gerhard Stöbich <deb2001-github@yahoo.de>
Reviewed-By: James M Snell <jasnell@gmail.com>
KEM encapsulation produces separate ciphertext and shared-secret
allocations. The existing DeriveBitsJob path packs both values into an
intermediate buffer, then copies them again into separate buffers.
Instead, this uses a dedicated KEMEncapsulateJob to retain both outputs
across the worker boundary and convert each directly through ByteSource.
This removes the intermediate allocation and at least one complete round
of copies.
Signed-off-by: Filip Skokan <panva.ip@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/64553
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Yagiz Nizipli <yagiz@nizipli.com>
The import.meta example for new vm.SourceTextModule() in vm.md does not
run as written. It fails in two separate ways.
First, the constructor is missing the context: contextifiedObject
option, so the module evaluates in the top context where secret is not
defined, and the snippet throws ReferenceError: secret is not defined.
Second, the trailing note suggests replacing meta.prop = {} with
vm.runInContext('{}', contextifiedObject), but '{}' is parsed as an
empty block and evaluates to undefined. That makes the following
Object.getPrototypeOf(import.meta.prop) throw TypeError. Wrapping it as
'({})' returns an object, which is what the note intends.
This adds the context option and corrects the suggested replacement to
'({})' in both the mjs and cjs variants.
Fixes: https://github.com/nodejs/node/issues/64076
Signed-off-by: Muhammad Zeeshan <61280174+zeeshan56656@users.noreply.github.com>
PR-URL: https://github.com/nodejs/node/pull/64112
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>
The THROW_IF_INSUFFICIENT_PERMISSIONS and
ASYNC_THROW_IF_INSUFFICIENT_PERMISSIONS macros called
ThrowAccessDenied/AsyncThrowAccessDenied unconditionally and only
guarded the `return` with `warning_only()`.
ERR_ACCESS_DENIED_IF_INSUFFICIENT_PERMISSIONS had no `warning_only()`
guard at all — it always set the access-denied error and returned.
As a result, running with `--permission-audit` still produced
ERR_ACCESS_DENIED on any denied operation (fs, net, child_process,
worker, addon, ffi, inspector, wasi), defeating the audit-only purpose
of the flag.
Guard the denied-error path behind `!warning_only()` in all three
macros. In audit mode, the diagnostics-channel message is published
(already done in Permission::is_scope_granted) and execution continues;
in enforce mode (`--permission`), behavior is unchanged — the error is
raised and the call returns.
The tests cover both the direct (top-level) call and an `eval()`-wrapped
call: the direct call exercises the normal script path, and the
`eval()`-wrapped call exercises the V8 script-context boundary (the
diagnostics subscriber is registered in the outer module context while
the denied operation runs inside an eval'd string).
Refs: https://github.com/nodejs/node/commit/9ddd1a9c27c253f46d587a8c906ccd83417b4606
Signed-off-by: Adrian Estrada <edsadr@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/64426
Reviewed-By: Rafael Gonzaga <rafael.nunu@hotmail.com>
Apply util.styleText to the `node --help` output for improved visual
hierarchy and scannability. Styling is applied only when the output
stream supports color (detected via util.styleText's built-in
shouldColorize), so piped/redirected output and NO_COLOR remain plain
text with no behavior change.
- Bold: Usage lines, Options:, Environment variables: headers
- Bold green: CLI option names (e.g. --inspect, -e, --eval)
- Bold magenta: environment variable names (e.g. NODE_PATH)
- Dim: "(currently set)" annotation
- Blue underline: documentation URL
Column-width math uses unstyled string lengths to preserve alignment
regardless of styling. The "(currently set)" annotation is styled via
post-processing after layout to avoid breaking the fold() width
calculation.
Signed-off-by: Adrian Estrada <edsadr@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/64484
Reviewed-By: Juan José Arboleda <soyjuanarbol@gmail.com>
Reviewed-By: Rafael Gonzaga <rafael.nunu@hotmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Gürgün Dayıoğlu <hey@gurgun.day>