Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms.
CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 ------------- Commit messages: - Remove extra whitespace. - Merge - Initial revision. Changes: https://git.openjdk.java.net/jdk/pull/2463/files Webrev: https://webrevs.openjdk.java.net/?repo=jdk&pr=2463&range=00 Issue: https://bugs.openjdk.java.net/browse/JDK-8259709 Stats: 55 lines in 5 files changed: 50 ins; 0 del; 5 mod Patch: https://git.openjdk.java.net/jdk/pull/2463.diff Fetch: git fetch https://git.openjdk.java.net/jdk pull/2463/head:pull/2463 PR: https://git.openjdk.java.net/jdk/pull/2463 |
On Mon, 8 Feb 2021 20:46:41 GMT, Sean Mullan <[hidden email]> wrote:
> Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms. > > CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 > Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 Marked as reviewed by rhalade (Reviewer). ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
In reply to this post by Sean Mullan-2
On Mon, 8 Feb 2021 20:46:41 GMT, Sean Mullan <[hidden email]> wrote:
> Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms. > > CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 > Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 Change looks good. ------------- Marked as reviewed by weijun (Reviewer). PR: https://git.openjdk.java.net/jdk/pull/2463 |
On Tue, 9 Feb 2021 21:04:00 GMT, Weijun Wang <[hidden email]> wrote:
>> Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms. >> >> CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 >> Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 > > Change looks good. All test changes are about re-enable disabled algorithms. Do we have a test on ensuring disabled algorithms are indeed disabled? ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
In reply to this post by Sean Mullan-2
On Mon, 8 Feb 2021 20:46:41 GMT, Sean Mullan <[hidden email]> wrote:
> Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms. > > CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 > Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 test/lib/jdk/test/lib/security/SecurityUtils.java line 78: > 76: * part of the algorithm URI. > 77: */ > 78: public static void removeAlgsFromDSigPolicy(List<String> algs) { How about using `String... algs` as arguments? ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
On Mon, 22 Feb 2021 03:42:23 GMT, Weijun Wang <[hidden email]> wrote:
>> Please review this change to disable XML signatures that use SHA-1 based digest or signature algorithms. SHA-1 is weak and is not a recommended algorithm for digital signatures. This will improve out of the box security by restricting XML signatures that use SHA-1 algorithms. >> >> CSR: https://bugs.openjdk.java.net/browse/JDK-8261246 >> Release Note: https://bugs.openjdk.java.net/browse/JDK-8261364 > > test/lib/jdk/test/lib/security/SecurityUtils.java line 78: > >> 76: * part of the algorithm URI. >> 77: */ >> 78: public static void removeAlgsFromDSigPolicy(List<String> algs) { > > How about using `String... algs` as arguments? Yes, that is nicer. ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
In reply to this post by Weijun Wang-2
On Fri, 19 Feb 2021 22:36:24 GMT, Weijun Wang <[hidden email]> wrote:
> All test changes are about re-enabling disabled algorithms. Do we have a test on ensuring disabled algorithms are indeed disabled? How about we set "org.jcp.xml.dsig.secureValidation" to false everywhere in the existing tests and add a new dedicated test to check for disabled algorithms/key sizes etc. That is what test/jdk/javax/xml/crypto/dsig/SecureValidationPolicy.java does, see this code block on lines 65-69: for (String alg : restrictedAlgs) { if (!Policy.restrictAlg(alg)) { throw new Exception(alg + " alg not restricted"); } } ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
On Wed, 24 Feb 2021 22:02:45 GMT, Sean Mullan <[hidden email]> wrote:
> > All test changes are about re-enabling disabled algorithms. Do we have a test on ensuring disabled algorithms are indeed disabled? How about we set "org.jcp.xml.dsig.secureValidation" to false everywhere in the existing tests and add a new dedicated test to check for disabled algorithms/key sizes etc. > > That is what test/jdk/javax/xml/crypto/dsig/SecureValidationPolicy.java does, see this code block on lines 65-69: > > ``` > for (String alg : restrictedAlgs) { > if (!Policy.restrictAlg(alg)) { > throw new Exception(alg + " alg not restricted"); > } > } > ``` This is only about checking the parsing function of the Policy class. I would be more confident if an actual validation call is made. I have a test on PSS at https://github.com/openjdk/jdk/blob/a79df58e0ad0b19aa8e0611cc55f5628383c2950/test/jdk/javax/xml/crypto/dsig/SecureValidation.java. Maybe I can enhance it to contain more algorithms. ------------- PR: https://git.openjdk.java.net/jdk/pull/2463 |
Free forum by Nabble | Edit this page |