[Hardy] Update from alpha 3 to alpha 4 broke ssh connections to openbsd boxes

Bug #187473 reported by Caroline Ford
4
Affects Status Importance Assigned to Milestone
openssh (Ubuntu)
Invalid
Undecided
Unassigned

Bug Description

Binary package hint: openssh-client

Running hardy alpha 3 I could ssh to a remote machine. Having fully updated I cannot any longer.

openssh-client 1:4.7p1-2

Same connection worked before I updated this evening, and still works on a different machine.
------
Update:

It works okay connecting to up-to-date linux machines. It appears to be an incompatibility connected to openbsd servers.

Revision history for this message
Caroline Ford (secretlondon) wrote :

secret@celery:~$ ssh -vvv eco.li -l secretlondon
OpenSSH_4.7p1 Debian-2, OpenSSL 0.9.8g 19 Oct 2007
debug1: Reading configuration data /etc/ssh/ssh_config
debug1: Applying options for *
debug2: ssh_connect: needpriv 0
debug1: Connecting to eco.li [195.10.223.77] port 22.
debug1: Connection established.
debug1: identity file /home/secret/.ssh/identity type -1
debug1: identity file /home/secret/.ssh/id_rsa type -1
debug1: identity file /home/secret/.ssh/id_dsa type -1
debug1: Remote protocol version 1.99, remote software version OpenSSH_4.3
debug1: match: OpenSSH_4.3 pat OpenSSH*
debug1: Enabling compatibility mode for protocol 2.0
debug1: Local version string SSH-2.0-OpenSSH_4.7p1 Debian-2
debug2: fd 3 setting O_NONBLOCK
debug1: SSH2_MSG_KEXINIT sent
Read from socket failed: Connection reset by peer

Revision history for this message
Caroline Ford (secretlondon) wrote :

iptables was also upgraded and seems to have had an error:

Setting up iptables (1.3.8.0debian1-1ubuntu2) ...
I/O error : Attempt to load network entity http://www.w3.org/TR/xhtml1/DTD/xhtml-lat1.ent
I/O error : Attempt to load network entity http://www.w3.org/TR/xhtml1/DTD/xhtml-special.ent
I/O error : Attempt to load network entity http://www.w3.org/TR/xhtml1/DTD/xhtml-symbol.ent

(from /var/log/apt/term.log)

Revision history for this message
Caroline Ford (secretlondon) wrote :
Revision history for this message
Caroline Ford (secretlondon) wrote :

It never gets past debug1: SSH2_MSG_KEXINIT sent

Revision history for this message
Caroline Ford (secretlondon) wrote :
Revision history for this message
Caroline Ford (secretlondon) wrote :

I've tried sshing to a different machine and got a lot further:

debug1: Authentications that can continue: publickey
debug1: Next authentication method: publickey
debug1: Trying private key: /home/secret/.ssh/identity
debug1: Trying private key: /home/secret/.ssh/id_rsa
debug1: Trying private key: /home/secret/.ssh/id_dsa
debug1: No more authentication methods to try.
Permission denied (publickey).
secret@celery:~$

I don't have a valid key so it's behaving as expected

Revision history for this message
Caroline Ford (secretlondon) wrote :

Failing ssh connection is to an openbsd box running remote software version OpenSSH_4.3
Okay ssh connection is to a linux box running remote software version OpenSSH_4.6p1 Debian-5ubuntu0.1

Revision history for this message
Caroline Ford (secretlondon) wrote :

http://kerneltrap.org/mailarchive/openbsd-misc/2007/4/24/148492

Thread on openbsd-misc referring to the same problem on a gutsy machine.

http://kerneltrap.org/mailarchive/openbsd-misc/2007/4/24/148531

We have a workround:

"I told it to keep state on incoming ssh connections"
"What I had before was allow ssh traffic in non-statefully"

With those changes to the server it now works. I am told the server was

"Only when running pf"
"And not having a fully stateful firewall"

Hope this helps.

description: updated
description: updated
Revision history for this message
Chuck Short (zulcss) wrote :

Are you still having this problem?

Thanks
chuck

Changed in openssh:
status: New → Incomplete
Revision history for this message
Caroline Ford (secretlondon) wrote :

No - but the server I connect to is using the workround I mentioned above.

Revision history for this message
Steve Langasek (vorlon) wrote :

It's not clear to me that there is an Ubuntu bug here - clearly the firewall on the server wasn't allowing through some of the packets associated with the connection, but those packets constitute legitimate traffic, so isn't this a bug on the server side - with the "workaround" actually a correct fix for a misconfiguration? Is there an explanation somewhere of why the behavior of the ssh client should be considered wrong/buggy?

Revision history for this message
Caroline Ford (secretlondon) wrote : Re: [Bug 187473] Re: [Hardy] Update from alpha 3 to alpha 4 broke ssh connections to openbsd boxes

It seems to be only versions of Debian and Ubuntu that have this problem.

Revision history for this message
Jayson Rowe (jayson.rowe) wrote :

We are closing this bug report because it lacks the information we need to investigate the problem, as described in the previous comments. Please reopen it if you can give us the missing information, and don't hesitate to submit bug reports in the future. To reopen the bug report you can click on the current status, under the Status column, and change the Status back to "New". Thanks again!

Changed in openssh:
status: Incomplete → Invalid
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.