Bug 1373836
Summary: | Remove RC4 cipher and questionable MACs enabled by default from OpenSSH | ||
---|---|---|---|
Product: | Red Hat Enterprise Linux 6 | Reporter: | Jakub Svoboda <jsvoboda> |
Component: | openssh | Assignee: | Jakub Jelen <jjelen> |
Status: | CLOSED ERRATA | QA Contact: | Stefan Dordevic <sdordevi> |
Severity: | unspecified | Docs Contact: | Mirek Jahoda <mjahoda> |
Priority: | unspecified | ||
Version: | 6.8 | CC: | ahecox, ajb, bressers, jjelen, leonard-rh-bugzilla, mdshaikh, nmavrogi, phil, sdordevi, thibaut.pouzet, tmraz, toracat, vdanen, yozone |
Target Milestone: | rc | ||
Target Release: | --- | ||
Hardware: | Unspecified | ||
OS: | Unspecified | ||
Whiteboard: | |||
Fixed In Version: | openssh-5.3p1-120.el6 | Doc Type: | Deprecated Functionality |
Doc Text: |
Deprecated algorithms in *OpenSSH*: RC4, hmac-md5, and hmac-md5-96
With this update, the `arcfour256`, `arcfour128`, `arcfour` ciphers and the `hmac-md5`, `hmac-md5-96` Method Authentication Code (MAC) algorithms are deprecated. Note that this change does not affect any existing server configuration.
The system administrator can enable these deprecated algorithms by editing the `ssh_config` file, for example:
Host legacy
Ciphers arcfour
MACs hmac-md5
To completely restore the deprecated functionality, add the following snippet to the `/etc/ssh/ssh_config` file:
Ciphers aes128-ctr,aes192-ctr,aes256-ctr,arcfour256,arcfour128,aes128-cbc,3des-cbc,blowfish-cbc,cast128-cbc,aes192-cbc,aes256-cbc,arcfour,rijndael-cbc@lysator.liu.se
MACs hmac-md5,hmac-sha1,umac-64@openssh.com,hmac-sha2-256,hmac-sha2-512,hmac-ripemd160,hmac-ripemd160@openssh.com,hmac-sha1-96,hmac-md5-96
|
Story Points: | --- |
Clone Of: | Environment: | ||
Last Closed: | 2017-03-21 10:02:38 UTC | Type: | Bug |
Regression: | --- | Mount Type: | --- |
Documentation: | --- | CRM: | |
Verified Versions: | Category: | --- | |
oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |
Cloudforms Team: | --- | Target Upstream Version: | |
Embargoed: | |||
Bug Depends On: | |||
Bug Blocks: | 1335911 |
Description
Jakub Svoboda
2016-09-07 08:54:40 UTC
Well ... my bad. You are right. It is a bug. The hmac-md5-96 should not be removed from the server list. Only from the default client list. I will need to fix this. The test was obviously right. Sorry for the confusion. *** Bug 1417264 has been marked as a duplicate of this bug. *** Probably best to only leave around sha2-512 and sha2-256 enabled by default, and sha1 as a legacy option for HMACs. Not just MD5 is considered broken, but also RIPEMD. See http://csrc.nist.gov/groups/ST/toolkit/documents/shs/hash_standards_comments.pdf . Note that this is a NIST statement from 2004, that is THIRTEEN YEARS AGO. Things have only deteriorated since then. I'm not so sure the hype against CBC is warranted. With proper padding CBC is perfectly usable. Counter modes like GCM and CTR have their own issues, like injecting known state (albeit hashed/crypted) which should not be brushed aside so easily. Such known state injection might well be successfully leveraged by future cryptanalitic progress. So personally I'm not entirely convinced GCM is that much more secure (or even more secure at all) than (properly implemented) CBC. 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://rhn.redhat.com/errata/RHSA-2017-0641.html |