X2Go Bug report logs - #1618
x2go on Ubuntu MATE 22.04.3 LTS host & client reorders key presses

Reply or subscribe to this bug.

Toggle useless messages

Message #5 received at submit@bugs.x2go.org (full text, mbox, reply):

From: Christian Kothe <christiankothe@gmail.com>

To: submit@bugs.x2go.org

Subject: x2go on Ubuntu MATE 22.04.3 LTS host & client reorders key presses

Date: Sun, 17 Dec 2023 10:28:35 -0800

[Message part 1 (text/plain, inline)]

Package: x2goserver
Version: 4.1.0.3

Testing against: x2goclient 4.1.2.2  (note: since it's not clear whether
the problem is with the client or the server, I'm reposting the same issue
also for the x2goclient package to make sure it doesn't fall through the
cracks).

We're finding that, when typing rapidly on an x2go connection between two
stock Ubuntu MATE 22.04.3 installations with quite up-to-date x2go client &
server packages, the order of keystrokes is quite frequently scrambled
(maybe 1 in 20 to 1 in 100 keystrokes), which makes touch typing an
exercise in frustration.

Obviously there's always the question of whether the bug is perhaps with
the user, although I've been touch-typing for 25 years and the typo rate
(always key swaps) is extremely unusual for me. I also haven't seen that
with the exact same client installation (versions, OS, machine, WiFi)
against another MATE server-side setup (also using x2goserver 4.1.0.3, but
instead MATE 20.04.6 LTS, running on AWS, and in a similar location and
with similar server-side pings) if that's in any way helpful.

The connections are not particularly slow, but on the client side there's
generally a WiFi hop involved. Client-side speedtest.net results: 45Mbps
down / 22Mbps up, ping 21ms up & 204ms down. Server-side speedtest.net
results: 740Mbps down / 925 Mbps up, ping 6ms up and 7ms down.

Regards,
Christian Kothe

[Message part 2 (text/html, inline)]

Message #10 received at 1618@bugs.x2go.org (full text, mbox, reply):

From: Christian Kothe <christiankothe@gmail.com>

To: 1618@bugs.x2go.org

Subject: Additional information to aid reproducibility:

Date: Sun, 11 Feb 2024 07:33:06 -0800

[Message part 1 (text/plain, inline)]

The bug can be reproduced by running a test script that simulates
successive key presses in sequence at high (touch-typing) rates. The
attached screenshot shows the local copy of that script in the top half and
the key presses as they appear in a terminal window on the remote machine
in the bottom half. The annotations show that there is on average one
reordering every 7-8 closely timed keystrokes.

[image: testcase-annotated.png]

The script code is (you may have to run pip install pyautogui to install
the dependency into your python interpreter):
import time
import pyautogui

time.sleep(5)
for k in range(10):
    pyautogui.press(list('0123456789abcdefghijklmnopqrstuvwxyz'))
    pyautogui.press('enter')
    time.sleep(1)

Note that the editor is PyCharm (a popular Java-based IDE), which is a bit
more CPU-intensive than some other software during editing, so this could
well have something to do with it. If you want to test it with that,
PyCharm is freely available from: https://www.jetbrains.com/pycharm/

Thanks for your attention,
Christian

[Message part 2 (text/html, inline)]

[testcase-annotated.png (image/png, inline)]

Send a report that this bug log contains spam.