Bug 26758 - mbedtls new security issue CVE-2020-10932
Summary: mbedtls new security issue CVE-2020-10932
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: Security (show other bugs)
Version: 7
Hardware: All Linux
Priority: Normal normal
Target Milestone: ---
Assignee: QA Team
QA Contact: Sec team
URL:
Whiteboard: MGA7-64-OK
Keywords: advisory, validated_update
Depends on:
Blocks:
 
Reported: 2020-06-09 22:24 CEST by David Walser
Modified: 2020-06-16 09:46 CEST (History)
5 users (show)

See Also:
Source RPM: mbedtls-2.16.5-1.mga7.src.rpm
CVE:
Status comment:


Attachments

David Walser 2020-06-09 22:24:39 CEST

Whiteboard: (none) => MGA7TOO
Assignee: bugsquad => rverschelde

Comment 1 Rémi Verschelde 2020-06-09 22:58:44 CEST
Weird, I committed 2.16.6 back on April 21, 2020, but it seems I did not submit it? Will have a look.
Comment 2 Rémi Verschelde 2020-06-09 23:04:25 CEST
Fixed in Cauldron with mbedtls-2.16.6-1.mga8.

Pushed mbedtls-2.16.6-1.mga7 to Mageia 7 core/updates_testing.

Advisory:
=========

Updated mbedtls packages fix security vulnerability

  Fix side channel in ECC code that allowed an adversary with access to precise
  enough timing and memory access information (typically an untrusted operating
  system attacking a secure enclave) to fully recover an ECDSA private key.
  (CVE-2020-10932)
  
  Fix a potentially remotely exploitable buffer overread in a DTLS client when
  parsing the Hello Verify Request message.

References:
 - https://tls.mbed.org/tech-updates/releases/mbedtls-2.16.6-and-2.7.15-released
 - https://tls.mbed.org/tech-updates/security-advisories/mbedtls-security-advisory-2020-04

RPMs in core/updates_testing:
=============================

lib64mbedcrypto3-2.16.6-1.mga7
lib64mbedtls12-2.16.6-1.mga7
lib64mbedtls-devel-2.16.6-1.mga7
lib64mbedx509_0-2.16.6-1.mga7
mbedtls-2.16.6-1.mga7

SRPM in core/updates_testing:
=============================

mbedtls-2.16.6-1.mga7

CC: (none) => rverschelde
Assignee: rverschelde => qa-bugs

Comment 3 Herman Viaene 2020-06-10 14:53:23 CEST
MGA7-64 Plasma on Lenovo B50
No instalaation issues.
Ref bug 2625 Comment2for testing.
So started godot from CLI in a third desktop (QARepo and MCC in second as I usually do), then switched back to Desktop 1 to read the bug ref'ed above. Came back to Desktop 3, noticed that the window of godot lost all its contents, clicked on that window and the whole machine frooze, no interaction with mouse or keyboard possible. Left it there for 15 min.,then had to power the laptop off. Rebooted without problems.

CC: (none) => herman.viaene

Comment 4 Rémi Verschelde 2020-06-10 15:37:45 CEST
That would be an issue with Godot itself, not related to mbedtls a priori.
Lenovo B50 has relatively low end graphics so it might not be able to handle Godot's OpenGL 3.3 requirements fully.
Nicolas Lécureuil 2020-06-10 22:16:54 CEST

CC: (none) => mageia
Keywords: (none) => advisory

Comment 5 Len Lawrence 2020-06-14 21:05:41 CEST
Picking up the baton Herman - nvidia GTX970.
Ran the self-test suite before updating.  Passed all 27 tests.

Clean update of the five packages.
$ mbedtls-selftest
....
  TIMING test #3 (hardclock / get_timer): passed
  Executed 27 test suites
  [ All tests PASS ]

Installed godot.
$ godot
Godot Engine v3.2.1.stable.mageia - https://godotengine.org
OpenGL ES 3.0 Renderer: GeForce GTX 970/PCIe/SSE2
 
Project is missing: /home/lcl/godot/project.godot
Editing project: /home/lcl/godot (::home::lcl::godot)
Godot Engine v3.2.1.stable.mageia - https://godotengine.org
lcl@difda:~ $ OpenGL ES 3.0 Renderer: GeForce GTX 970/PCIe/SSE2
 
ERROR: set_path: Another resource is loaded from path 'res://art/enemyFlyingAlt_1.png' (possible cyclic resource inclusion).
   At: core/resource.cpp:81.

This presented a gui which allowed games to be downloaded and installed.  Chose one and installed it in a new folder and ran the editor.  That looked perfectly functional but left it there and played the demo of "Dodge the Creeps".

Hopefully this is enough for an OK.

CC: (none) => tarazed25
Whiteboard: MGA7TOO => MGA7TOO MGA7-64-OK

Len Lawrence 2020-06-14 21:07:32 CEST

Whiteboard: MGA7TOO MGA7-64-OK => MGA7TOO

Comment 6 Len Lawrence 2020-06-14 21:08:45 CEST
Hmm.  Not network tested.
Comment 7 Len Lawrence 2020-06-14 21:36:01 CEST
Ran  godot under strace.  Downloaded and installed a game and ran it from the editor.
$ grep mbedtls godot.trace
openat(AT_FDCWD, "/lib64/libmbedtls.so.12", O_RDONLY|O_CLOEXEC) = 3
openat(AT_FDCWD, "/usr/lib64/libmbedtls.so.2.16.6", O_RDONLY) = 3
openat(AT_FDCWD, "/usr/lib64/libmbedtls.so.2.16.6", O_RDONLY) = 4

$ grep -i tls godot.trace
shows a lot of messages like this:
clone(child_stack=0x7fb7e9539ef0, flags=CLONE_VM|CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD|CLONE_SYSVSEM|CLONE_SETTLS|CLONE_PARENT_SETTID|CLONE_CHILD_CLEARTID, parent_tid=[32422], tls=0x7fb7e953a700, child_tidptr=0x7fb7e953a9d0) = 32422

No idea if any of this intersects the area(s) concerned in the issues reported upstream.
Comment 8 Rémi Verschelde 2020-06-15 10:40:36 CEST
Yeah that's a good test, any download made from Godot uses HTTPS and thus mbedTLS.

Whiteboard: MGA7TOO => MGA7TOO MGA7-64-OK

Comment 9 Len Lawrence 2020-06-15 11:46:57 CEST
Thanks Rémi - validating.

Keywords: (none) => validated_update
CC: (none) => sysadmin-bugs

Comment 10 Len Lawrence 2020-06-16 08:32:30 CEST
What is the protocol for validating a QA update flagged as MGA7TOO?  What about Cauldron?
Nicolas Lécureuil 2020-06-16 09:01:51 CEST

Whiteboard: MGA7TOO MGA7-64-OK => MGA7-64-OK
Version: Cauldron => 7

Comment 11 Mageia Robot 2020-06-16 09:46:20 CEST
An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGASA-2020-0265.html

Resolution: (none) => FIXED
Status: NEW => RESOLVED


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