Bug 1533667 - Add: standard algorithm names for all ECC signature types
Summary: Add: standard algorithm names for all ECC signature types
Keywords:
Status: CLOSED ERRATA
Alias: None
Product: Red Hat Enterprise Linux 7
Classification: Red Hat
Component: jss
Version: 7.5
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: rc
: ---
Assignee: Christina Fu
QA Contact: ipa-qe
URL:
Whiteboard:
Depends On: 1533666
Blocks:
TreeView+ depends on / blocked
 
Reported: 2018-01-11 22:37 UTC by Matthew Harmsen
Modified: 2018-04-10 17:57 UTC (History)
12 users (show)

Fixed In Version: jss-4.4.0-11.el7
Doc Type: No Doc Update
Doc Text:
undefined
Clone Of: 1533666
Environment:
Last Closed: 2018-04-10 17:56:52 UTC
Target Upstream Version:
Embargoed:


Attachments (Terms of Use)


Links
System ID Private Priority Status Summary Last Updated
Mozilla Foundation 589158 0 None None None 2018-01-11 22:37:59 UTC
Red Hat Product Errata RHBA-2018:0958 0 None None None 2018-04-10 17:57:58 UTC

Description Matthew Harmsen 2018-01-11 22:37:59 UTC
+++ This bug was initially created as a clone of Bug #1533666 +++

In https://bugzilla.mozilla.org/show_bug.cgi?id=589158, David Stutzman Created attachment 467756 [details] - "correct patch showing + for lines instead of -"

JSS uses SHAxwithEC as their naming for EC Signature algorithms but Sun's "Standard Algorithm Names" are SHAxwithECDSA.  I was using Sun's CertPath API to attempt to chain some EC certificates that had SHA384withECDSA signatures and the building of the CertPath threw an exception because it couldn't find a provider that supported SHA384withECDSA.  The attached patch just adds the 4 aliases for the standard Sun algorithm names to the JSS provider.

Adding these aliases should enable JSS's ECC functionality to be used in more places throughout the entire Java library as it would now support the algorithm names that the library uses.

--- Additional comment from Matthew Harmsen on 2018-01-11 17:33:23 EST ---

Wan-Teh Chang approved this patch:

r=wtc.

It would be nice if we could make SHA1withECDSA
the algorithm name and make SHA1withEC an alias.
Would that break any code?

--- Additional comment from Matthew Harmsen on 2018-01-11 17:34:02 EST ---

David Stutzman replied:

As far as I could tell, the way you guys are creating the provider now, it's reporting the main names of SHAxwithEC.  After I added the new aliases I queried the Provider for it's Provider.Services and called getAlgorithm() and it doesn't explicity list the new ones (or any aliases).  It seems to only show the initial algorithm name.  I don't know if there's a way to get all the aliases for a particular Provider.Service. http://download.oracle.com/javase/6/docs/api/java/security/Provider.Service.html says: "Optionally, it also includes a list of alternate algorithm names for this service (aliases) and attributes, which are a map of (name, value) String pairs," but there doesn't appear to be any getter for that info.

http://download.oracle.com/javase/6/docs/technotes/guides/security/crypto/HowToImplAProvider.html#Aliases  Not a whole lot of helpful info here, but that's the official word on aliases.

On the topic of breaking existing code by changing defaults:

Looking at it with this patch applied if you call  Signature.getInstance("SHA256withEC"), it will work.  If you call Signature.getInstance("SHA256withECDSA"), it will work.

If you change the other name to be the default, again either call above will work.  

The only thing I can see breaking is if someone queries the Provider programatically using Provider.getServices() and then accessing the Provider.Service objects to get names and is expecting to match on the old string (SHAxwithEC) at that point because that won't list aliases.

Most likely it wouldn't be a huge deal because I suspect that not many people are using JSS/NSS for ECC at this time, but that's your call.

Comment 3 Christina Fu 2018-01-12 02:16:59 UTC
pushed:

https://bugzilla.mozilla.org/show_bug.cgi?id=589158#c6

Comment 5 Sumedh Sidhaye 2018-01-31 17:10:59 UTC
Hi Christina,

Do I need to perform any tests other than ensuring ECC stuff is working for this bug, if yes could you please mention it here?

Comment 6 Christina Fu 2018-02-01 19:35:33 UTC
just make sure it doesn't break ECC should be good enough.

Comment 7 Sumedh Sidhaye 2018-02-02 11:32:41 UTC
Build used for verification:

root@csqa4-guest01 ~ # rpm -qi jss
Name        : jss
Version     : 4.4.0
Release     : 11.el7
Architecture: x86_64
Install Date: Monday 29 January 2018 10:37:07 PM EST
Group       : System Environment/Libraries
Size        : 1029926
License     : MPLv1.1 or GPLv2+ or LGPLv2+
Signature   : RSA/SHA256, Monday 22 January 2018 07:02:01 PM EST, Key ID 199e2f91fd431d51
Source RPM  : jss-4.4.0-11.el7.src.rpm
Build Date  : Monday 22 January 2018 06:17:27 PM EST
Build Host  : x86-040.build.eng.bos.redhat.com
Relocations : (not relocatable)
Packager    : Red Hat, Inc. <http://bugzilla.redhat.com/bugzilla>
Vendor      : Red Hat, Inc.
URL         : http://www.mozilla.org/projects/security/pki/jss/
Summary     : Java Security Services (JSS)

All the certificate system (ECC) operations are working as expected.

Hence marking this bug verified.

Comment 10 errata-xmlrpc 2018-04-10 17:56:52 UTC
Since the problem described in this bug report should be
resolved in a recent advisory, it has been closed with a
resolution of ERRATA.

For information on the advisory, and where to find the updated
files, follow the link below.

If the solution does not work for you, open a new bug report.

https://access.redhat.com/errata/RHBA-2018:0958


Note You need to log in before you can comment on or make changes to this bug.