Bug 2439 - New sha256-base64 SSH Fingerprints in openssh-6.8
Summary: New sha256-base64 SSH Fingerprints in openssh-6.8
Status: CLOSED WONTFIX
Alias: None
Product: Portable OpenSSH
Classification: Unclassified
Component: Miscellaneous (show other bugs)
Version: 6.9p1
Hardware: Other Linux
: P5 enhancement
Assignee: Assigned to nobody
URL:
Keywords:
Depends on:
Blocks: V_8_0
  Show dependency treegraph
 
Reported: 2015-08-06 00:26 AEST by Jakub Jelen
Modified: 2021-04-23 14:58 AEST (History)
1 user (show)

See Also:


Attachments
Allow more fingerprint algorithms on client side (11.49 KB, patch)
2015-08-07 18:27 AEST, Jakub Jelen
no flags Details | Diff

Note You need to log in before you can comment on or make changes to this bug.
Description Jakub Jelen 2015-08-06 00:26:32 AEST
Based on our Fedora bug [1] I started investigation what is up to with the new Fingerprint hashes in openssh-6.8. I found one inconsistency and a usability problem.


 1) First of all manual pages mention that: 

> Valid options are: “md5” and “sha256”.

but both config parser and all tools accepts ALL digests defined in "digest-{openssl,glibc}.c" in array digests[], which contains much more of them and which do not have any support and can lead to misunderstanding. I propose to strip the list according to documentation. But it collides a bit with the other proposal:


 2) As I stated in previously mentioned bugzilla, it would be great to have the way to show more Fingerprint types, since the most of the servers still provide only the old fingerprint (and for some years probably will). Also it is not preferable to stuck with old md5 as default. You can admit, that users can always do

 $ ssh server -oFingerprintHash=md5

but it is probably too much for users if they really want to verify fingerpring provided through other channel.

My proposal is to add ability to provide a list of digest that will be printed (not only one) and as a transition default use both available: "sha256,md5".


I don't have a patch yet, but if there would be some idea how can we make the transition more smooth, feel free to comment.

[1] https://bugzilla.redhat.com/show_bug.cgi?id=1249626
Comment 1 Jakub Jelen 2015-08-07 18:27:32 AEST
Created attachment 2681 [details]
Allow more fingerprint algorithms on client side

Adding patch for client side, which introduces new default "sha256 md5". For internal operations and logging, only the first one is used, but for interaction with user (adding new host), both fingerprints are printed.

From my point of view, there is no need to do the same on the server side, since there is no interaction with user.
Comment 2 Damien Miller 2016-02-26 14:44:25 AEDT
Retarget to openssh-7.3
Comment 3 Damien Miller 2016-02-26 14:47:21 AEDT
Retarget to openssh-7.3
Comment 4 Damien Miller 2016-07-22 14:10:54 AEST
retarget unfinished bugs to next release
Comment 5 Damien Miller 2016-07-22 14:14:38 AEST
retarget unfinished bugs to next release
Comment 6 Damien Miller 2016-07-22 14:15:47 AEST
retarget unfinished bugs to next release
Comment 7 Damien Miller 2016-07-22 14:17:12 AEST
retarget unfinished bugs to next release
Comment 8 Damien Miller 2016-12-16 14:31:21 AEDT
OpenSSH 7.4 release is closing; punt the bugs to 7.5
Comment 9 Damien Miller 2017-06-30 13:43:19 AEST
Move incomplete bugs to openssh-7.6 target since 7.5 shipped a while back.

To calibrate expectations, there's little chance all of these are going to make 7.6.
Comment 10 Damien Miller 2017-06-30 13:44:22 AEST
remove 7.5 target
Comment 11 Damien Miller 2018-04-06 13:12:13 AEST
Move to OpenSSH 7.8 tracking bug
Comment 12 Damien Miller 2018-08-10 11:37:58 AEST
Retarget remaining bugs planned for 7.8 release to 7.9
Comment 13 Damien Miller 2018-08-10 11:38:25 AEST
Retarget remaining bugs planned for 7.8 release to 7.9
Comment 14 Damien Miller 2018-10-19 17:13:42 AEDT
Retarget unfinished bugs to OpenSSH 8.0
Comment 15 Damien Miller 2018-10-19 17:14:48 AEDT
Retarget unfinished bugs to OpenSSH 8.0
Comment 16 Damien Miller 2018-10-19 17:15:50 AEDT
Retarget unfinished bugs to OpenSSH 8.0
Comment 17 Damien Miller 2019-02-22 14:21:09 AEDT
I don't think I want to move forward with this change - the sha256 signatures are a lot more ubiquitous and accepted than they were in 2016 and I want to hasten the day when OpenSSH and its dependencies can compile without any MD5 support at all.
Comment 18 Damien Miller 2021-04-23 14:58:49 AEST
closing resolved bugs as of 8.6p1 release