Bug 2250 - SOCKS5 should return "NO ACCEPTABLE METHODS" instead of nothing
Summary: SOCKS5 should return "NO ACCEPTABLE METHODS" instead of nothing
Status: NEW
Alias: None
Product: Portable OpenSSH
Classification: Unclassified
Component: ssh (show other bugs)
Version: -current
Hardware: All All
: P5 enhancement
Assignee: Darren Tucker
URL:
Keywords:
Depends on: 2417
Blocks:
  Show dependency tree / graph
 
Reported: 2014-06-26 21:12 AEST by Gunter Grodotzki
Modified: 2016-06-17 14:52 AEST (History)
3 users (show)

See Also:

Attachments
Return "NO ACCEPTABLE METHODS" if we don't find the method we are looking for (1.26 KB, patch)
2015-06-05 14:07 AEST, Darren Tucker 		no flags Details | Diff
View All Add an attachment (proposed patch, testcase, etc.)

Note You need to log in before you can comment on or make changes to this bug.
Description Gunter Grodotzki 2014-06-26 21:12:56 AEST 
In: http://www.openbsd.org/cgi-bin/cvsweb/src/usr.bin/ssh/channels.c?rev=1.332;content-type=text%2Fplain
Search for: "SSH_SOCKS5_NOAUTH not found"

If a connection to SSH-SOCKS5 is established and a unsupported auth method is sent, SSH will not reply anything.

According to http://tools.ietf.org/html/rfc1928 the server could (it does not state if the server MUST or SHOULD, hence "could" ;) ) send simply X'FF' NO ACCEPTABLE METHODS - with this a client can do appropriate handling.

Would like to hear your thoughts if I am misinterpreting rfc1928, or I should just suck it up and handle cases like this with my client.
Comment 1 Darren Tucker 2015-06-05 13:50:30 AEST 
My reading of RFC1928 agrees with yours.  I'll take a look at this after the upcoming 6.9 release is out the door.
Comment 2 Darren Tucker 2015-06-05 14:07:41 AEST 
Created attachment 2644 [details]
Return "NO ACCEPTABLE METHODS" if we don't find the method we are looking for

I think this will do it but I don't have an easy way to test it.
Comment 3 Jonas Berlin 2015-06-25 22:08:13 AEST 
Will the buffered bytes be sent even when the method returns -1 which causes the caller to call chan_mark_dead(); ?
Comment 4 Jonas Berlin 2015-06-25 22:18:37 AEST 
Attempted to implement error reporting for a few more error situations in bug #2417.
Comment 5 Damien Miller 2015-08-11 22:59:12 AEST 
Retarget pending bugs to openssh-7.1
Comment 6 Damien Miller 2016-02-26 14:44:29 AEDT 
Retarget to openssh-7.3
Comment 7 Damien Miller 2016-02-26 14:47:16 AEDT 
Retarget to openssh-7.3
Comment 8 Damien Miller 2016-06-17 14:52:46 AEST 
Remove from openssh-7.3 list until the blocking bug is fixed