Bug 7813 - Client reports "Internal error in SSH." for some connection errors
Summary: Client reports "Internal error in SSH." for some connection errors
Status: NEW
Alias: None
Product: ThinLinc
Classification: Unclassified
Component: Client (show other bugs)
Version: trunk
Hardware: PC Unknown
: P2 Normal
Target Milestone: MediumPrio
Assignee: Bugzilla mail exporter
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2021-12-23 15:51 CET by Pierre Ossman
Modified: 2022-01-10 08:19 CET (History)
0 users

See Also:
Acceptance Criteria:


Attachments

Description Pierre Ossman cendio 2021-12-23 15:51:01 CET
Something was amiss with our ThinLinc server and I had problems connection, getting this in the log:

> 2021-12-23T15:40:54: SSH command: /opt/thinlinc/lib/tlclient/ssh -N -o GlobalKnownHostsFile=/dev/null -o UserKnownHostsFile=/dev/null -o UpdateHostKeys=yes -o PubkeyAuthentication=no -o CheckHostIP=no -o NumberOfPasswordPrompts=1 -o HostKeyAlgorithms=ssh-rsa,ecdsa-sha2-nistp256 ossman@tl.cendio.se -p 22 -L 35451:127.0.0.1:5914 -R 5045:127.0.0.1:48135 -R 5046:127.0.0.1:43777 -R 5044:127.0.0.1:37063 -R 5043:127.0.0.1:43121 thinlinc-login dummy
> 2021-12-23T15:40:54: SSH pid is 2352355
> 2021-12-23T15:40:54: ssh[E]: kex_exchange_identification: read: Connection reset by peer
> 2021-12-23T15:40:54: ssh[E]: SYSTEM ERROR: 104

This is a network error, so we should report it as such to the user. But instead I got "Internal error in SSH.", which was unexpected.

Looking at the code we except network errors in this stage to report "CONNECT ERROR", not "SYSTEM ERROR". However this code is from 2009, so things might have changed in OpenSSH and that assumption is no longer correct.

(or it might have always been wrong and we never caught this particular case before)

I have no idea what was wrong with the server, so I don't know how to properly reproduce this unfortunately.

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