To comply with RFC 7301, make TLS servers send a fatal alert during the
TLS handshake if both the client and the server are configured to use
ALPN and if the server does not support any of the protocols advertised
by the client.
This affects HTTP/2 servers. Until now, applications could intercept the
'unknownProtocol' event when the client either did not advertise any
protocols or if the list of protocols advertised by the client did not
include HTTP/2 (or HTTP/1.1 if allowHTTP1 was true). With this change,
only the first case can be handled, and the 'unknownProtocol' event will
not be emitted in the second case because the TLS handshake fails and no
secure connection is established.
PR-URL: https://github.com/nodejs/node/pull/44031
Reviewed-By: Paolo Insogna <paolo@cowtech.it>
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Rafael Gonzaga <rafael.nunu@hotmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
The getCipher() returns a tuple that includes protocol version string.
This string refers to the minimum protocol version string, as per
documentation. What is missing is a reference to the documentation
where to get the negotiated cipher for the socket connection and
a clearer example.
Fixes: https://github.com/nodejs/node/issues/43406
PR-URL: https://github.com/nodejs/node/pull/44086
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Some core modules can be loaded with or without the `node:` prefix.
Using the prefix disambiguates which specifiers refer to core modules.
This commit updates the docs to use the prefix everywhere a core module
is referenced.
PR-URL: https://github.com/nodejs/node/pull/42752
Fixes: https://github.com/nodejs/node/issues/38343
Reviewed-By: Matteo Collina <matteo.collina@gmail.com>
Reviewed-By: Mestery <mestery@protonmail.com>
Reviewed-By: Darshan Sen <raisinten@gmail.com>
Reviewed-By: Akhil Marsonya <akhil.marsonya27@gmail.com>
Reviewed-By: Rafael Gonzaga <rafael.nunu@hotmail.com>
Reviewed-By: Mohammed Keyvanzadeh <mohammadkeyvanzade94@gmail.com>
Reviewed-By: Geoffrey Booth <webadmin@geoffreybooth.com>
Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Notable Changes:
Deprecations and Removals:
- (SEMVER-MAJOR) fs: runtime deprecate string coercion in `fs.write`,
`fs.writeFileSync`
(Livia Medeiros) (https://github.com/nodejs/node/pull/42607)
- (SEMVER-MAJOR) dns: remove `dns.lookup` and `dnsPromises.lookup`
options type coercion
(Antoine du Hamel) (https://github.com/nodejs/node/pull/41431)
- (SEMVER-MAJOR) process: runtime deprecate multipleResolves
(Benjamin Gruenbaum) (https://github.com/nodejs/node/pull/41896)
- (SEMVER-MAJOR) stream: remove thenable support (Robert Nagy)
(https://github.com/nodejs/node/pull/40773)
- (SEMVER-MAJOR) tls: move tls.parseCertString to end-of-life
(Tobias Nießen) (https://github.com/nodejs/node/pull/41479)
fetch (experimental):
An experimental fetch API is available on the global scope by default.
The implementation is based upon https://undici.nodejs.org/#/,
an HTTP/1.1 client written for Node.js by contributors to the project.
Through this addition, the following globals are made available: `fetch`
, `FormData`, `Headers`, `Request`, `Response`.
Disable this API with the `--no-experimental-fetch` command-line flag.
Contributed by Michaël Zasso in https://github.com/nodejs/node/pull/41811.
HTTP Timeouts:
`server.headersTimeout`, which limits the amount of time the parser will
wait to receive the complete HTTP headers, is now set to `60000` (60
seconds) by default.
`server.requestTimeout`, which sets the timeout value in milliseconds
for receiving the entire request from the client, is now set to `300000`
(5 minutes) by default.
If these timeouts expire, the server responds with status 408 without
forwarding the request to the request listener and then closes the
connection.
Both timeouts must be set to a non-zero value to protect against
potential Denial-of-Service attacks in case the server is deployed
without a reverse proxy in front.
Contributed by Paolo Insogna in https://github.com/nodejs/node/pull/41263.
Test Runner module (experimental):
The `node:test` module facilitates the creation of JavaScript tests that
report results in TAP format. This module is only available under the
`node:` scheme.
Contributed by Colin Ihrig in https://github.com/nodejs/node/pull/42325.
Toolchain and Compiler Upgrades:
- Prebuilt binaries for Linux are now built on Red Hat Enterprise Linux
(RHEL) 8 and are compatible with Linux distributions based on glibc
2.28 or later, for example, Debian 10, RHEL 8, Ubuntu 20.04.
- Prebuilt binaries for macOS now require macOS 10.15 or later.
- For AIX the minimum supported architecture has been raised from Power
7 to Power 8.
Prebuilt binaries for 32-bit Windows will initially not be available due
to issues building the V8 dependency in Node.js. We hope to restore
32-bit Windows binaries for Node.js 18 with a future V8 update.
Node.js does not support running on operating systems that are no longer
supported by their vendor. For operating systems where their vendor has
planned to end support earlier than April 2025, such as Windows 8.1
(January 2023) and Windows Server 2012 R2 (October 2023), support for
Node.js 18 will end at the earlier date.
Full details about the supported toolchains and compilers are documented
in the Node.js `BUILDING.md` file.
Contributed by Richard Lau in https://github.com/nodejs/node/pull/42292,
https://github.com/nodejs/node/pull/42604 and https://github.com/nodejs/node/pull/42659
, and Michaël Zasso in https://github.com/nodejs/node/pull/42105 and
https://github.com/nodejs/node/pull/42666.
V8 10.1:
The V8 engine is updated to version 10.1, which is part of Chromium 101.
Compared to the version included in Node.js 17.9.0, the following new
features are included:
- The `findLast` and `findLastIndex` array methods.
- Improvements to the `Intl.Locale` API.
- The `Intl.supportedValuesOf` function.
- Improved performance of class fields and private class methods (the
initialization of them is now as fast as ordinary property stores).
The data format returned by the serialization API (`v8.serialize(value)`)
has changed, and cannot be deserialized by earlier versions of Node.js.
On the other hand, it is still possible to deserialize the previous
format, as the API is backwards-compatible.
Contributed by Michaël Zasso in https://github.com/nodejs/node/pull/42657.
Web Streams API (experimental):
Node.js now exposes the experimental implementation of the Web Streams
API on the global scope. This means the following APIs are now globally
available:
- `ReadableStream`, `ReadableStreamDefaultReader`,
`ReadableStreamBYOBReader`, `ReadableStreamBYOBRequest`,
`ReadableByteStreamController`, `ReadableStreamDefaultController`,
`TransformStream`, `TransformStreamDefaultController`, `WritableStream`,
`WritableStreamDefaultWriter`, `WritableStreamDefaultController`,
`ByteLengthQueuingStrategy`, `CountQueuingStrategy`, `TextEncoderStream`,
`TextDecoderStream`, `CompressionStream`, `DecompressionStream`.
Contributed James Snell in https://github.com/nodejs/node/pull/39062,
and Antoine du Hamel in https://github.com/nodejs/node/pull/42225.
Other Notable Changes:
- (SEMVER-MAJOR) buffer: expose Blob as a global
(James M Snell) (https://github.com/nodejs/node/pull/41270)
- (SEMVER-MAJOR) child\_process: improve argument validation
(Rich Trott) (https://github.com/nodejs/node/pull/41305)
- doc: add RafaelGSS to collaborators
(RafaelGSS) (https://github.com/nodejs/node/pull/42718)
- (SEMVER-MAJOR) http: make TCP noDelay enabled by default
(Paolo Insogna) (https://github.com/nodejs/node/pull/42163)
- (SEMVER-MAJOR) net: make `server.address()` return an integer for
`family`
(Antoine du Hamel) (https://github.com/nodejs/node/pull/41431)
- (SEMVER-MAJOR) worker: expose BroadcastChannel as a global
(James M Snell) (https://github.com/nodejs/node/pull/41271)
- (SEMVER-MAJOR) worker: graduate BroadcastChannel to supported
(James M Snell) (https://github.com/nodejs/node/pull/41271)
Semver-Major Commits:
- (SEMVER-MAJOR) assert,util: compare RegExp.lastIndex while using deep
equal checks
(Ruben Bridgewater) (https://github.com/nodejs/node/pull/41020)
- (SEMVER-MAJOR) buffer: refactor `byteLength` to remove outdated
optimizations
(Rongjian Zhang) (https://github.com/nodejs/node/pull/38545)
- (SEMVER-MAJOR) buffer: expose Blob as a global
(James M Snell) (https://github.com/nodejs/node/pull/41270)
- (SEMVER-MAJOR) buffer: graduate Blob from experimental
(James M Snell) (https://github.com/nodejs/node/pull/41270)
- (SEMVER-MAJOR) build: make x86 Windows support temporarily
experimental
(Michaël Zasso) (https://github.com/nodejs/node/pull/42666)
- (SEMVER-MAJOR) build: bump macOS deployment target to 10.15
(Richard Lau) (https://github.com/nodejs/node/pull/42292)
- (SEMVER-MAJOR) build: downgrade Windows 8.1 and server 2012 R2 to
experimental
(Michaël Zasso) (https://github.com/nodejs/node/pull/42105)
- (SEMVER-MAJOR) child\_process: improve argument validation
(Rich Trott) (https://github.com/nodejs/node/pull/41305)
- (SEMVER-MAJOR) cluster: make `kill` to be just `process.kill`
(Bar Admoni) (https://github.com/nodejs/node/pull/34312)
- (SEMVER-MAJOR) crypto: cleanup validation
(Mohammed Keyvanzadeh) (https://github.com/nodejs/node/pull/39841)
- (SEMVER-MAJOR) crypto: prettify othername in PrintGeneralName
(Tobias Nießen) (https://github.com/nodejs/node/pull/42123)
- (SEMVER-MAJOR) crypto: fix X509Certificate toLegacyObject
(Tobias Nießen) (https://github.com/nodejs/node/pull/42124)
- (SEMVER-MAJOR) crypto: use RFC2253 format in PrintGeneralName
(Tobias Nießen) (https://github.com/nodejs/node/pull/42002)
- (SEMVER-MAJOR) crypto: change default check(Host|Email) behavior
(Tobias Nießen) (https://github.com/nodejs/node/pull/41600)
- (SEMVER-MAJOR) deps: V8: cherry-pick semver-major commits from 10.2
(Michaël Zasso) (https://github.com/nodejs/node/pull/42657)
- (SEMVER-MAJOR) deps: update V8 to 10.1.124.6
(Michaël Zasso) (https://github.com/nodejs/node/pull/42657)
- (SEMVER-MAJOR) deps: update V8 to 9.8.177.9
(Michaël Zasso) (https://github.com/nodejs/node/pull/41610)
- (SEMVER-MAJOR) deps: update V8 to 9.7.106.18
(Michaël Zasso) (https://github.com/nodejs/node/pull/40907)
- (SEMVER-MAJOR) dns: remove `dns.lookup` and `dnsPromises.lookup`
options type coercion
(Antoine du Hamel) (https://github.com/nodejs/node/pull/41431)
- (SEMVER-MAJOR) doc: update minimum glibc requirements for Linux
(Richard Lau) (https://github.com/nodejs/node/pull/42659)
- (SEMVER-MAJOR) doc: update AIX minimum supported arch
(Richard Lau) (https://github.com/nodejs/node/pull/42604)
- (SEMVER-MAJOR) fs: runtime deprecate string coercion in `fs.write`,
`fs.writeFileSync`
(Livia Medeiros) (https://github.com/nodejs/node/pull/42607)
- (SEMVER-MAJOR) http: refactor headersTimeout and requestTimeout logic
(Paolo Insogna) (https://github.com/nodejs/node/pull/41263)
- (SEMVER-MAJOR) http: make TCP noDelay enabled by default
(Paolo Insogna) (https://github.com/nodejs/node/pull/42163)
- (SEMVER-MAJOR) lib: enable fetch by default
(Michaël Zasso) (https://github.com/nodejs/node/pull/41811)
- (SEMVER-MAJOR) lib: replace validator and error
(Mohammed Keyvanzadeh) (https://github.com/nodejs/node/pull/41678)
- (SEMVER-MAJOR) module,repl: support 'node:'-only core modules
(Colin Ihrig) (https://github.com/nodejs/node/pull/42325)
- (SEMVER-MAJOR) net: make `server.address()` return an integer for
`family`
(Antoine du Hamel) (https://github.com/nodejs/node/pull/41431)
- (SEMVER-MAJOR) process: disallow some uses of Object.defineProperty()
on process.env
(Himself65) (https://github.com/nodejs/node/pull/28006)
- (SEMVER-MAJOR) process: runtime deprecate multipleResolves
(Benjamin Gruenbaum) (https://github.com/nodejs/node/pull/41896)
- (SEMVER-MAJOR) readline: fix question still called after closed
(Xuguang Mei) (https://github.com/nodejs/node/pull/42464)
- (SEMVER-MAJOR) stream: remove thenable support
(Robert Nagy) (https://github.com/nodejs/node/pull/40773)
- (SEMVER-MAJOR) stream: expose web streams globals, remove runtime
experimental warning
(Antoine du Hamel) (https://github.com/nodejs/node/pull/42225)
- (SEMVER-MAJOR) stream: need to cleanup event listeners if last stream
is readable
(Xuguang Mei) (https://github.com/nodejs/node/pull/41954)
- (SEMVER-MAJOR) stream: revert revert `map` spec compliance
(Benjamin Gruenbaum) (https://github.com/nodejs/node/pull/41933)
- (SEMVER-MAJOR) stream: throw invalid arg type from End Of Stream
(Jithil P Ponnan) (https://github.com/nodejs/node/pull/41766)
- (SEMVER-MAJOR) stream: don't emit finish after destroy
(Robert Nagy) (https://github.com/nodejs/node/pull/40852)
- (SEMVER-MAJOR) stream: add errored and closed props
(Robert Nagy) (https://github.com/nodejs/node/pull/40696)
- (SEMVER-MAJOR) test: add initial test module
(Colin Ihrig) (https://github.com/nodejs/node/pull/42325)
- (SEMVER-MAJOR) timers: refactor internal classes to ES2015 syntax
(Rabbit) (https://github.com/nodejs/node/pull/37408)
- (SEMVER-MAJOR) tls: represent registeredID numerically always
(Tobias Nießen) (https://github.com/nodejs/node/pull/41561)
- (SEMVER-MAJOR) tls: move tls.parseCertString to end-of-life
(Tobias Nießen) (https://github.com/nodejs/node/pull/41479)
- (SEMVER-MAJOR) url: throw on NULL in IPv6 hostname
(Rich Trott) (https://github.com/nodejs/node/pull/42313)
- (SEMVER-MAJOR) v8: make v8.writeHeapSnapshot() error codes consistent
(Darshan Sen) (https://github.com/nodejs/node/pull/42577)
- (SEMVER-MAJOR) v8: make writeHeapSnapshot throw if fopen fails
(Antonio Román) (https://github.com/nodejs/node/pull/41373)
- (SEMVER-MAJOR) worker: expose BroadcastChannel as a global
(James M Snell) (https://github.com/nodejs/node/pull/41271)
- (SEMVER-MAJOR) worker: graduate BroadcastChannel to supported
(James M Snell) (https://github.com/nodejs/node/pull/41271)
PR-URL: https://github.com/nodejs/node/pull/42262
This is not a function and should not have a return type.
PR-URL: https://github.com/nodejs/node/pull/42647
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
This sentence implies that AES-128 is preferred over AES-256 because
of a related-key attack from 2009. However, that attack by Alex
Biryukov, Orr Dunkelman, Nathan Keller, Dmitry Khovratovich, and Adi
Shamir, while impressive, is only effective against variants of
AES-256 with a reduced number of rounds and it requires related keys.
This means that the attack is not effective against AES-256 as it is
used within TLS.
(AES-128 is still often preferred over AES-256 simply because it is
believed to be sufficiently secure and because it is faster.)
PR-URL: https://github.com/nodejs/node/pull/42578
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Mestery <mestery@protonmail.com>
Reviewed-By: Mohammed Keyvanzadeh <mohammadkeyvanzade94@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/42063
Refs: https://github.com/nodejs/node/issues/42059
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Mestery <mestery@protonmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Michael Dawson <midawson@redhat.com>
Reviewed-By: Harshitha K P <harshitha014@gmail.com>
Reviewed-By: Juan José Arboleda <soyjuanarbol@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Spelling out numbers makes this part easier to read. Also remove the
unnecessary word 'last'.
PR-URL: https://github.com/nodejs/node/pull/41778
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Mestery <mestery@protonmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Nitzan Uziely <linkgoron@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/41649
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <rlau@redhat.com>
Reviewed-By: Darshan Sen <raisinten@gmail.com>
This statement is misleading in that it says "key generation is
expensive". ECDHE key generation (over the elliptic curves that are
commonly used for TLS) is insanely fast compared to most other types
of key generation.
This statement is irrelevant for TLS 1.3, which requires (EC)DHE.
Even if this statement is somewhat true for TLS 1.2, it does not
justify discouraging the use of (EC)DHE.
PR-URL: https://github.com/nodejs/node/pull/41528
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
This statement was objectively false. Clients usually only need to
generate and/or own a private key if the server sends a
CertificateRequest during the TLS handshake, which is not a common case.
PR-URL: https://github.com/nodejs/node/pull/41505
Reviewed-By: Filip Skokan <panva.ip@gmail.com>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Michael Dawson <midawson@redhat.com>
PR-URL: https://github.com/nodejs/node/pull/41477
Refs: https://github.com/nodejs/node/pull/24358
Reviewed-By: Colin Ihrig <cjihrig@gmail.com>
Reviewed-By: Richard Lau <rlau@redhat.com>
Reviewed-By: Tierney Cyren <hello@bnb.im>
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
This is a security release.
Notable changes:
Improper handling of URI Subject Alternative Names (Medium)(CVE-2021-44531)
- Accepting arbitrary Subject Alternative Name (SAN) types, unless a PKI
is specifically defined to use a particular SAN type, can result in
bypassing name-constrained intermediates. Node.js was accepting URI SAN
types, which PKIs are often not defined to use. Additionally, when a
protocol allows URI SANs, Node.js did not match the URI correctly.
- Versions of Node.js with the fix for this disable the URI SAN type when
checking a certificate against a hostname. This behavior can be
reverted through the `--security-revert` command-line option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44531
Certificate Verification Bypass via String Injection (Medium)(CVE-2021-44532)
- Node.js converts SANs (Subject Alternative Names) to a string format.
It uses this string to check peer certificates against hostnames when
validating connections. The string format was subject to an injection
vulnerability when name constraints were used within a certificate
chain, allowing the bypass of these name constraints.
- Versions of Node.js with the fix for this escape SANs containing the
problematic characters in order to prevent the injection. This
behavior can be reverted through the `--security-revert` command-line
option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44532
Incorrect handling of certificate subject and issuer fields (Medium)(CVE-2021-44533)
- Node.js did not handle multi-value Relative Distinguished Names
correctly. Attackers could craft certificate subjects containing a
single-value Relative Distinguished Name that would be interpreted as a
multi-value Relative Distinguished Name, for example, in order to inject
a Common Name that would allow bypassing the certificate subject
verification.
- Affected versions of Node.js do not accept multi-value Relative
Distinguished Names and are thus not vulnerable to such attacks
themselves. However, third-party code that uses node's ambiguous
presentation of certificate subjects may be vulnerable.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44533
Prototype pollution via `console.table` properties (Low)(CVE-2022-21824)
- Due to the formatting logic of the `console.table()` function it was
not safe to allow user controlled input to be passed to the `properties`
parameter while simultaneously passing a plain object with at least one
property as the first parameter, which could be `__proto__`. The
prototype pollution has very limited control, in that it only allows an
empty string to be assigned numerical keys of the object prototype.
- Versions of Node.js with the fix for this use a null protoype for the
object these properties are being assigned to.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-21824
PR-URL: https://github.com/nodejs-private/node-private/pull/311
This is a security release.
Notable changes:
Improper handling of URI Subject Alternative Names (Medium)(CVE-2021-44531)
- Accepting arbitrary Subject Alternative Name (SAN) types, unless a PKI
is specifically defined to use a particular SAN type, can result in
bypassing name-constrained intermediates. Node.js was accepting URI SAN
types, which PKIs are often not defined to use. Additionally, when a
protocol allows URI SANs, Node.js did not match the URI correctly.
- Versions of Node.js with the fix for this disable the URI SAN type when
checking a certificate against a hostname. This behavior can be
reverted through the `--security-revert` command-line option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44531
Certificate Verification Bypass via String Injection (Medium)(CVE-2021-44532)
- Node.js converts SANs (Subject Alternative Names) to a string format.
It uses this string to check peer certificates against hostnames when
validating connections. The string format was subject to an injection
vulnerability when name constraints were used within a certificate
chain, allowing the bypass of these name constraints.
- Versions of Node.js with the fix for this escape SANs containing the
problematic characters in order to prevent the injection. This
behavior can be reverted through the `--security-revert` command-line
option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44532
Incorrect handling of certificate subject and issuer fields (Medium)(CVE-2021-44533)
- Node.js did not handle multi-value Relative Distinguished Names
correctly. Attackers could craft certificate subjects containing a
single-value Relative Distinguished Name that would be interpreted as a
multi-value Relative Distinguished Name, for example, in order to inject
a Common Name that would allow bypassing the certificate subject
verification.
- Affected versions of Node.js do not accept multi-value Relative
Distinguished Names and are thus not vulnerable to such attacks
themselves. However, third-party code that uses node's ambiguous
presentation of certificate subjects may be vulnerable.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44533
Prototype pollution via `console.table` properties (Low)(CVE-2022-21824)
- Due to the formatting logic of the `console.table()` function it was
not safe to allow user controlled input to be passed to the `properties`
parameter while simultaneously passing a plain object with at least one
property as the first parameter, which could be `__proto__`. The
prototype pollution has very limited control, in that it only allows an
empty string to be assigned numerical keys of the object prototype.
- Versions of Node.js with the fix for this use a null protoype for the
object these properties are being assigned to.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-21824
PR-URL: https://github.com/nodejs-private/node-private/pull/312
This is a security release.
Notable changes:
Improper handling of URI Subject Alternative Names (Medium)(CVE-2021-44531)
- Accepting arbitrary Subject Alternative Name (SAN) types, unless a PKI
is specifically defined to use a particular SAN type, can result in
bypassing name-constrained intermediates. Node.js was accepting URI SAN
types, which PKIs are often not defined to use. Additionally, when a
protocol allows URI SANs, Node.js did not match the URI correctly.
- Versions of Node.js with the fix for this disable the URI SAN type when
checking a certificate against a hostname. This behavior can be
reverted through the `--security-revert` command-line option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44531
Certificate Verification Bypass via String Injection (Medium)(CVE-2021-44532)
- Node.js converts SANs (Subject Alternative Names) to a string format.
It uses this string to check peer certificates against hostnames when
validating connections. The string format was subject to an injection
vulnerability when name constraints were used within a certificate
chain, allowing the bypass of these name constraints.
- Versions of Node.js with the fix for this escape SANs containing the
problematic characters in order to prevent the injection. This
behavior can be reverted through the `--security-revert` command-line
option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44532
Incorrect handling of certificate subject and issuer fields (Medium)(CVE-2021-44533)
- Node.js did not handle multi-value Relative Distinguished Names
correctly. Attackers could craft certificate subjects containing a
single-value Relative Distinguished Name that would be interpreted as a
multi-value Relative Distinguished Name, for example, in order to inject
a Common Name that would allow bypassing the certificate subject
verification.
- Affected versions of Node.js do not accept multi-value Relative
Distinguished Names and are thus not vulnerable to such attacks
themselves. However, third-party code that uses node's ambiguous
presentation of certificate subjects may be vulnerable.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44533
Prototype pollution via `console.table` properties (Low)(CVE-2022-21824)
- Due to the formatting logic of the `console.table()` function it was
not safe to allow user controlled input to be passed to the `properties`
parameter while simultaneously passing a plain object with at least one
property as the first parameter, which could be `__proto__`. The
prototype pollution has very limited control, in that it only allows an
empty string to be assigned numerical keys of the object prototype.
- Versions of Node.js with the fix for this use a null protoype for the
object these properties are being assigned to.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-21824
PR-URL: https://github.com/nodejs-private/node-private/pull/310
This is a security release.
Notable changes:
Improper handling of URI Subject Alternative Names (Medium)(CVE-2021-44531)
- Accepting arbitrary Subject Alternative Name (SAN) types, unless a PKI
is specifically defined to use a particular SAN type, can result in
bypassing name-constrained intermediates. Node.js was accepting URI SAN
types, which PKIs are often not defined to use. Additionally, when a
protocol allows URI SANs, Node.js did not match the URI correctly.
- Versions of Node.js with the fix for this disable the URI SAN type when
checking a certificate against a hostname. This behavior can be
reverted through the `--security-revert` command-line option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44531
Certificate Verification Bypass via String Injection (Medium)(CVE-2021-44532)
- Node.js converts SANs (Subject Alternative Names) to a string format.
It uses this string to check peer certificates against hostnames when
validating connections. The string format was subject to an injection
vulnerability when name constraints were used within a certificate
chain, allowing the bypass of these name constraints.
- Versions of Node.js with the fix for this escape SANs containing the
problematic characters in order to prevent the injection. This
behavior can be reverted through the `--security-revert` command-line
option.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44532
Incorrect handling of certificate subject and issuer fields (Medium)(CVE-2021-44533)
- Node.js did not handle multi-value Relative Distinguished Names
correctly. Attackers could craft certificate subjects containing a
single-value Relative Distinguished Name that would be interpreted as a
multi-value Relative Distinguished Name, for example, in order to inject
a Common Name that would allow bypassing the certificate subject
verification.
- Affected versions of Node.js do not accept multi-value Relative
Distinguished Names and are thus not vulnerable to such attacks
themselves. However, third-party code that uses node's ambiguous
presentation of certificate subjects may be vulnerable.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2021-44533
Prototype pollution via `console.table` properties (Low)(CVE-2022-21824)
- Due to the formatting logic of the `console.table()` function it was
not safe to allow user controlled input to be passed to the `properties`
parameter while simultaneously passing a plain object with at least one
property as the first parameter, which could be `__proto__`. The
prototype pollution has very limited control, in that it only allows an
empty string to be assigned numerical keys of the object prototype.
- Versions of Node.js with the fix for this use a null protoype for the
object these properties are being assigned to.
- More details will be available at
https://cve.mitre.org/cgi-bin/cvename.cgi?name=CVE-2022-21824
PR-URL: https://github.com/nodejs-private/node-private/pull/309
Previously, Node.js incorrectly accepted uniformResourceIdentifier (URI)
subject alternative names in checkServerIdentity regardless of the
application protocol. This was incorrect even in the most common cases.
For example, RFC 2818 specifies (and RFC 6125 confirms) that HTTP over
TLS only uses dNSName and iPAddress subject alternative names, but not
uniformResourceIdentifier subject alternative names.
Additionally, name constrained certificate authorities might not be
constrained to specific URIs, allowing them to issue certificates for
URIs that specify hosts that they would not be allowed to issue dNSName
certificates for.
Even for application protocols that make use of URI subject alternative
names (such as SIP, see RFC 5922), Node.js did not implement the
required checks correctly, for example, because checkServerIdentity
ignores the URI scheme.
As a side effect, this also fixes an edge case. When a hostname is not
an IP address and no dNSName subject alternative name exists, the
subject's Common Name should be considered even when an iPAddress
subject alternative name exists.
It remains possible for users to pass a custom checkServerIdentity
function to the TLS implementation in order to implement custom identity
verification logic.
This addresses CVE-2021-44531.
CVE-ID: CVE-2021-44531
PR-URL: https://github.com/nodejs-private/node-private/pull/300
Reviewed-By: Michael Dawson <midawson@redhat.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
These changes are in preparation for enabling a JSDoc lint rule.
PR-URL: https://github.com/nodejs/node/pull/41109
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Ruben Bridgewater <ruben@bridgewater.de>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/40959
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
Reviewed-By: Rich Trott <rtrott@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
PR-URL: https://github.com/nodejs/node/pull/39809
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Tobias Nießen <tniessen@tnie.de>
Reviewed-By: Filip Skokan <panva.ip@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Update mdast-util-to-markdown to 1.2.4 which reduces unnecessary
escaping of `_` and some other characters. Re-run markdown formatter.
PR-URL: https://github.com/nodejs/node/pull/40645
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
PR-URL: https://github.com/nodejs/node/pull/40403
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: Zijian Liu <lxxyxzj@gmail.com>
Reviewed-By: Michaël Zasso <targos@protonmail.com>
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
Main changes:
- Replace current HTML anchor generation to match
header anchor generation in Github markdown.
- Remove unnecessary double namespacing on generated anchors/links (E.G.
`esm.md#loaders` instead of `esm.md#esm_loaders`).
- Anchors/links are automatically prefixed with their respective modules
when concatenated for usage in `all.html`.
Benefits:
- All anchor links within and between markdown API docs actually work.
- Adding new anchor links no longer requires contributors to generate
the HTML docs first to look up the correct anchors.
- Anchors are much shorter.
- All previous anchor links are preserved by generating hidden legacy
anchors.
PR-URL: https://github.com/nodejs/node/pull/39304
Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com>
Minor changes mostly to improve compliance with our style guide.
PR-URL: https://github.com/nodejs/node/pull/38202
Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Luigi Pinca <luigipinca@gmail.com>
