Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add a new option to limit DH key size in tls connect #1831

Closed
wants to merge 3 commits into from

Conversation

shigeki
Copy link
Contributor

@shigeki shigeki commented May 29, 2015

This is the fix of logjam attack for a TLS client. It disable connecting to the TLS server with less than DH key size specified by a new option of minDHkeylen. The default is 1024 bits.

And it also introduces a new public API of TLSSocket.getEphemeralKeyInfo() to show the information of an ephemeral key type, name and length of TLS connection. I think it it good for users to know them. We should prepare future vulnerabilities not only DH but also ECDH.
Unfortunately openssl provided SSL_get_server_tmp_key() on only client so that this API returns an empty object on the server.

The last one is my home work from @bnoordhuis to output warning of setDHParams to console.trace().

Here is a list of logjam fix in openssl, Firefox and Chrome. I think iojs is also better to set minimum key length like this.

R= @bnoordhuis @indutny

@shigeki shigeki added crypto Issues and PRs related to the crypto subsystem. tls Issues and PRs related to the tls subsystem. semver-minor PRs that contain new features and should be released in the next minor version. labels May 29, 2015
@shigeki shigeki changed the title Fix logjam attack for tls client Add a new option to limit DH key size in tls connect against logjam attack May 29, 2015
@indutny
Copy link
Member

indutny commented May 29, 2015

Are you sure that we are actually vulnerable to the logjam attack? I always thought that we are not doing False Start, so it should not be possible to perform MiTM attack in this case.

@shigeki
Copy link
Contributor Author

shigeki commented May 29, 2015

False Start is the worst case to decrypt DATA in offline.
But the paper of https://weakdh.org/imperfect-forward-secrecy.pdf says

we can keep browser connections alive by sending TLS warning alerts

According the paper, 512 bit DH key can be solved about 90 secs for the server with built-in prime such as apache. The current TLS timeout of iojs is 120 secs which is longer than that.

@indutny
Copy link
Member

indutny commented May 29, 2015

@shigeki but this is not logjam anymore, right? I just don't want to say that we was vulnerable to it in the first place, especially if we wasn't.

@shigeki shigeki changed the title Add a new option to limit DH key size in tls connect against logjam attack Add a new option to limit DH key size in tls connect May 29, 2015
@shigeki
Copy link
Contributor Author

shigeki commented May 29, 2015

If the wording of logjam only covers export DH cipher, that is right. I did not care about that. I've just changed this PR title to remove the word of logjam. There is no word of logjam in the commit message.

@@ -347,6 +347,11 @@ Creates a new client connection to the given `port` and `host` (old API) or

- `session`: A `Buffer` instance, containing TLS session.

- `minDHkeylen`: Minimum key length of an ephemeral DH parameter to
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

A camel case (minDHKeyLen)? Or even minDHSize

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I take minDHSize as it is shorter and more legible.

@indutny
Copy link
Member

indutny commented May 29, 2015

A couple of nits, otherwise looking great! Good job, @shigeki !

@shigeki
Copy link
Contributor Author

shigeki commented May 30, 2015

@bnoordhuis @indutny Thanks for reviewing. According to your comments, I updated them. PTAL.

connection. It returns an empty object when the key exchange is not
ephemeral. As it is only supported on a client socket, it returns null
if this is called on a server socket. The type of 'DH' and 'ECDH' are
supported. The name property is only available in 'ECDH'.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Femto-nit: maybe reword this to:

The supported types are 'DH' and 'ECDH'.  The `name` property is only available with 'ECDH'.

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Fixed

@bnoordhuis
Copy link
Member

LGTM apart from a couple of nits. Can you rebase and run the CI before landing this? Thanks.

@shigeki shigeki force-pushed the fix_LogjamAttack_client branch 3 times, most recently from 8b16dfd to e4f02bd Compare June 1, 2015 16:37
@jasnell
Copy link
Member

jasnell commented Oct 16, 2015

@shigeki @bnoordhuis @indutny ... ping

@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

Sorry for my late response. This is a security enhancement of DH as the same way of Chrome/Firefox. It is not so much threat at the usual use but https://freedom-to-tinker.com/blog/haldermanheninger/how-is-nsa-breaking-so-much-crypto/ says that

Breaking a second 1024-bit prime would allow passive eavesdropping on connections to nearly 20% of the top million HTTPS websites. In other words, a one-time investment in massive computation would make it possible to eavesdrop on trillions of encrypted connections.

so that I think it is worth to have this in order to protect against the nation level attack.
The current patch is revised after the last LGTM and I'm not sure that whether I can land it or not without a new review.

@indutny
Copy link
Member

indutny commented Oct 16, 2015

This should be semver-major. It will definitely break code that was connecting to 1024-bit DH servers. Otherwise LGTM

@shigeki shigeki added semver-major PRs that contain breaking changes and should be released in the next major version. and removed semver-minor PRs that contain new features and should be released in the next minor version. labels Oct 16, 2015
@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

@indutny Thanks. I agree it. I've changed label. I rebase this, make tests again and land it after 5.0 start.

@Fishrock123
Copy link
Member

Note: semver-major will have to land before 5.0 or else be delayed until 6.0

@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

@Fishrock123 Thanks. The current master is already 5.0.0-pre so can I land this now?

@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

CI show the following 7 test failures but they are not related to this PR. Otherwise are green.
@Fishrock123 @rvagg Could you give me a timing to land semver-major?

  • ubuntu1404-32
    not ok 42 test-child-process-fork-net2.js
  • vs2013,win2008r2
    not ok 11 test-child-process-fork-regr-Segfault in node::StreamBase::GetFD #2847.js
  • vs2015,win2012r2
    not ok 223 test-http-pipeline-flood.js
  • ppcle-ubuntu1404
    not ok 904 test-debug-args.js
  • ppcbe-fedora20
    not ok 45 test-child-process-fork-regr-Segfault in node::StreamBase::GetFD #2847.js
    not ok 712 test-tick-processor.js
    not ok 904 test-debug-args.js

@Fishrock123
Copy link
Member

@shigeki v5.0 is set to cut probably next week, and semver-major's have been landing on master. If this PR is green as usual it can land on master at any time, but it will only make it into v5 if it lands soon. :)

(This also assumes this has had enough TSC review, which I think it has.)

shigeki pushed a commit that referenced this pull request Oct 16, 2015
Returns an object representing a type, name and size of an ephemeral
key exchange in a client connection. Currently only DHE and ECHE are
supported.

This api only works on on a client connection. When it is called on a
server connection, null is returned. When its key exchange is not
ephemeral, an empty object is returned.

PR-URL: #1831
Reviewed-By: indutny - Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
shigeki pushed a commit that referenced this pull request Oct 16, 2015
Add a new option to specifiy a minimum size of an ephemeral DH
parameter to accept a tls connection. Default is 1024 bit.

PR-URL: #1831
Reviewed-By: indutny - Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
shigeki pushed a commit that referenced this pull request Oct 16, 2015
To make it easy to figure out where the warning comes from.
Also fix style and variable name that was made in #1739.

PR-URL: #1831
Reviewed-By: indutny - Fedor Indutny <fedor.indutny@gmail.com>
Reviewed-By: bnoordhuis - Ben Noordhuis <info@bnoordhuis.nl>
@shigeki
Copy link
Contributor Author

shigeki commented Oct 16, 2015

@Fishrock123 Thanks.
Landed in 6d92eba, f72e178 and 0140e1b to the current master.

@shigeki shigeki closed this Oct 16, 2015
@rvagg rvagg mentioned this pull request Oct 21, 2015
rvagg added a commit to rvagg/io.js that referenced this pull request Oct 29, 2015
Notable changes:

* buffer: (Breaking) Removed both 'raw' and 'raws' encoding types from Buffer,
  these have been deprecated for a long time (Sakthipriyan Vairamani) nodejs#2859.
* console: (Breaking) Values reported by console.time() now have 3 decimals of
  accuracy added (Michaël Zasso) nodejs#3166.
* fs:
  - fs.readFile*(), fs.writeFile*(), and fs.appendFile*() now also accept a file
    descriptor as their first argument (Johannes Wüller) nodejs#3163.
  - (Breaking) In fs.readFile(), if an encoding is specified and the internal
    toString() fails the error is no longer thrown but is passed to the callback
    (Evan Lucas) nodejs#3485.
  - (Breaking) In fs.read() (using the fs.read(fd, length, position, encoding,
    callback) form), if the internal toString() fails the error is no longer
    thrown but is passed to the callback (Evan Lucas) nodejs#3503.
* http:
  - Fixed a bug where pipelined http requests would stall (Fedor Indutny) nodejs#3342.
  - (Breaking) When parsing HTTP, don't add duplicates of the following headers:
    Retry-After, ETag, Last-Modified, Server, Age, Expires. This is in addition
    to the following headers which already block duplicates: Content-Type,
    Content-Length, User-Agent, Referer, Host, Authorization,
    Proxy-Authorization, If-Modified-Since, If-Unmodified-Since, From, Location,
    Max-Forwards (James M Snell) nodejs#3090.
  - (Breaking) The callback argument to OutgoingMessage#setTimeout() must be a
    function or a TypeError is thrown (James M Snell) nodejs#3090.
  - (Breaking) HTTP methods and header names must now conform to the RFC 2616
    "token" rule, a list of allowed characters that excludes control characters
    and a number of separator characters. Specifically, methods and header names
    must now match /^[a-zA-Z0-9_!#$%&'*+.^`|~-]+$/ or a TypeError will be thrown
    (James M Snell) nodejs#2526.
* node:
  - (Breaking) Deprecated the _linklist module (Rich Trott) nodejs#3078.
  - (Breaking) Removed require.paths and require.registerExtension(), both had
    been previously set to throw Error when accessed
    (Sakthipriyan Vairamani) nodejs#2922.
* npm: Upgraded to version 3.3.6 from 2.14.7, see
  https://github.com/npm/npm/releases/tag/v3.3.6 for more details. This is a
  major version bump for npm and it has seen a significant amount of change.
  Please see the original npm v3.0.0 release notes for a list of major changes
  (Rebecca Turner) nodejs#3310.
* src: (Breaking) Bumped NODE_MODULE_VERSION to 47 from 46, this is necessary
  due to the V8 upgrade. Native add-ons will need to be recompiled
  (Rod Vagg) nodejs#3400.
* timers: Attempt to reuse the timer handle for setTimeout().unref(). This fixes
  a long-standing known issue where unrefed timers would perviously hold
  beforeExit open (Fedor Indutny) nodejs#3407.
* tls:
  - Added ALPN Support (Shigeki Ohtsu) nodejs#2564.
  - TLS options can now be passed in an object to createSecurePair()
    (Коренберг Марк) nodejs#2441.
  - (Breaking) The default minimum DH key size for tls.connect() is now 1024
    bits and a warning is shown when DH key size is less than 2048 bits. This a security consideration to prevent "logjam" attacks. A new minDHSize TLS
    option can be used to override the default. (Shigeki Ohtsu) nodejs#1831.
* util:
  - (Breaking) util.p() was deprecated for years, and has now been removed
    (Wyatt Preul) nodejs#3432.
  - (Breaking) util.inherits() can now work with ES6 classes. This is considered
    a breaking change because of potential subtle side-effects caused by a
    change from directly reassigning the prototype of the constructor using
    `ctor.prototype = Object.create(superCtor.prototype, { constructor: { ... } })`
    to using `Object.setPrototypeOf(ctor.prototype, superCtor.prototype)`
    (Michaël Zasso) nodejs#3455.
* v8: (Breaking) Upgraded to 4.6.85.25 from 4.5.103.35 (Ali Ijaz Sheikh) nodejs#3351.
  - Implements the spread operator, see
    https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Spread_operator
    for further information.
  - Implements new.target, see
    https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/new.target
    for further information.
* zlib: Decompression now throws on truncated input (e.g. unexpected end of
  file) (Yuval Brik) nodejs#2595.

PR-URL: nodejs#3466
rvagg added a commit that referenced this pull request Oct 29, 2015
Notable changes:

* buffer: (Breaking) Removed both 'raw' and 'raws' encoding types from Buffer,
  these have been deprecated for a long time (Sakthipriyan Vairamani) #2859.
* console: (Breaking) Values reported by console.time() now have 3 decimals of
  accuracy added (Michaël Zasso) #3166.
* fs:
  - fs.readFile*(), fs.writeFile*(), and fs.appendFile*() now also accept a file
    descriptor as their first argument (Johannes Wüller) #3163.
  - (Breaking) In fs.readFile(), if an encoding is specified and the internal
    toString() fails the error is no longer thrown but is passed to the callback
    (Evan Lucas) #3485.
  - (Breaking) In fs.read() (using the fs.read(fd, length, position, encoding,
    callback) form), if the internal toString() fails the error is no longer
    thrown but is passed to the callback (Evan Lucas) #3503.
* http:
  - Fixed a bug where pipelined http requests would stall (Fedor Indutny) #3342.
  - (Breaking) When parsing HTTP, don't add duplicates of the following headers:
    Retry-After, ETag, Last-Modified, Server, Age, Expires. This is in addition
    to the following headers which already block duplicates: Content-Type,
    Content-Length, User-Agent, Referer, Host, Authorization,
    Proxy-Authorization, If-Modified-Since, If-Unmodified-Since, From, Location,
    Max-Forwards (James M Snell) #3090.
  - (Breaking) The callback argument to OutgoingMessage#setTimeout() must be a
    function or a TypeError is thrown (James M Snell) #3090.
  - (Breaking) HTTP methods and header names must now conform to the RFC 2616
    "token" rule, a list of allowed characters that excludes control characters
    and a number of separator characters. Specifically, methods and header names
    must now match /^[a-zA-Z0-9_!#$%&'*+.^`|~-]+$/ or a TypeError will be thrown
    (James M Snell) #2526.
* node:
  - (Breaking) Deprecated the _linklist module (Rich Trott) #3078.
  - (Breaking) Removed require.paths and require.registerExtension(), both had
    been previously set to throw Error when accessed
    (Sakthipriyan Vairamani) #2922.
* npm: Upgraded to version 3.3.6 from 2.14.7, see
  https://github.com/npm/npm/releases/tag/v3.3.6 for more details. This is a
  major version bump for npm and it has seen a significant amount of change.
  Please see the original npm v3.0.0 release notes for a list of major changes
  (Rebecca Turner) #3310.
* src: (Breaking) Bumped NODE_MODULE_VERSION to 47 from 46, this is necessary
  due to the V8 upgrade. Native add-ons will need to be recompiled
  (Rod Vagg) #3400.
* timers: Attempt to reuse the timer handle for setTimeout().unref(). This fixes
  a long-standing known issue where unrefed timers would perviously hold
  beforeExit open (Fedor Indutny) #3407.
* tls:
  - Added ALPN Support (Shigeki Ohtsu) #2564.
  - TLS options can now be passed in an object to createSecurePair()
    (Коренберг Марк) #2441.
  - (Breaking) The default minimum DH key size for tls.connect() is now 1024
    bits and a warning is shown when DH key size is less than 2048 bits. This a security consideration to prevent "logjam" attacks. A new minDHSize TLS
    option can be used to override the default. (Shigeki Ohtsu) #1831.
* util:
  - (Breaking) util.p() was deprecated for years, and has now been removed
    (Wyatt Preul) #3432.
  - (Breaking) util.inherits() can now work with ES6 classes. This is considered
    a breaking change because of potential subtle side-effects caused by a
    change from directly reassigning the prototype of the constructor using
    `ctor.prototype = Object.create(superCtor.prototype, { constructor: { ... } })`
    to using `Object.setPrototypeOf(ctor.prototype, superCtor.prototype)`
    (Michaël Zasso) #3455.
* v8: (Breaking) Upgraded to 4.6.85.25 from 4.5.103.35 (Ali Ijaz Sheikh) #3351.
  - Implements the spread operator, see
    https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/Spread_operator
    for further information.
  - Implements new.target, see
    https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Operators/new.target
    for further information.
* zlib: Decompression now throws on truncated input (e.g. unexpected end of
  file) (Yuval Brik) #2595.

PR-URL: #3466
@Fishrock123 Fishrock123 added the semver-minor PRs that contain new features and should be released in the next minor version. label Oct 30, 2015
@Fishrock123
Copy link
Member

Tagging also as minor since one commit is just additive.

@MylesBorins
Copy link
Member

Just want to verify that f72e178 will not be landing in lts

@jasnell
Copy link
Member

jasnell commented Nov 17, 2015

@thealphanerd ... this one is a special case, a variation on this will land in v4.x, v0.12 and v0.10 as part of a security fix that landed in master but not in the branches.

@MylesBorins
Copy link
Member

@jasnell has the security fix landed yet? I could not find any of these commits in v4.x-staging

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
crypto Issues and PRs related to the crypto subsystem. semver-major PRs that contain breaking changes and should be released in the next major version. semver-minor PRs that contain new features and should be released in the next minor version. tls Issues and PRs related to the tls subsystem.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

None yet

7 participants