xpra icon
Bug tracker and wiki

Custom Query (2323 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (4 - 6 of 2323)

1 2 3 4 5 6 7 8 9 10 11 12
Ticket Resolution Summary Owner Reporter
#2589 invalid Socket exception: An existing connection was forcibly closed by the remote host (10054) Antoine Martin stdedos
Description

I know it's an old r24762 version, but maybe it is interesting.

Out of the blue, my shadow-session disconnected:

"Xpra-Python3-x86_64_4.0-r24762\xpra_cmd" shadow ssh://user@ip/0 --opengl=no --desktop-scaling=0.75 --min-speed=70 --webcam=no --speaker=off --microphone=off --pulseaudio=no

2020-02-11 21:56:05,024 Xpra GTK3 client version 4.0-r24762 64-bit
2020-02-11 21:56:05,028  running on Microsoft Windows 10
2020-02-11 21:56:05,088 Warning: failed to import opencv:
2020-02-11 21:56:05,088  No module named 'cv2'
2020-02-11 21:56:05,088  webcam forwarding is disabled
2020-02-11 21:56:05,877 GStreamer version 1.16.2 for Python 3.8.0 64-bit
2020-02-11 21:56:06,425 OpenGL_accelerate module loaded
2020-02-11 21:56:06,509 Using accelerated ArrayDatatype
2020-02-11 21:56:07,197 Warning: vendor 'Intel' is greylisted,
2020-02-11 21:56:07,197  you may want to turn off OpenGL if you encounter bugs
2020-02-11 21:56:07,533 OpenGL enabled with Intel(R) HD Graphics 4000
2020-02-11 21:56:07,777  keyboard settings: layout=us
2020-02-11 21:56:07,781  desktop size is 1600x900 with 1 screen:
2020-02-11 21:56:07,781   Default (423x238 mm - DPI: 96x96) workarea: 1600x860
2020-02-11 21:56:07,781     (Standard monitor types) Generic PnP Monitor (309x174 mm - DPI: 131x131)
2020-02-11 21:56:11,342 enabled remote logging
2020-02-11 21:56:11,350 Xpra GTK3 X11 server version 3.0.6-r25174 64-bit
2020-02-11 21:56:11,354  running on Linux Ubuntu 16.04 xenial
2020-02-11 21:56:11,374 Attached to 172.16.57.121:22
2020-02-11 21:56:11,378  (press Control-C to detach)
[....]
2020-02-11 21:25:54,551 server is not responding, drawing spinners over the windows
2020-02-11 21:25:55,762 Socket exception: An existing connection was forcibly closed by the remote host (10054)
2020-02-11 21:25:56,173 Warning: failed to query SO_DONTROUTE, SO_ERROR, SO_EXCLUSIVEADDRUSE, SO_KEEPALIVE, SO_LINGER, SO_OOBINLINE, SO_RCVBUF, SO_RCVTIMEO, SO_REUSEADDR, SO_SNDBUF, SO_SNDTIMEO, SO_TYPE
2020-02-11 21:25:56,215 Connection lost

And server has nothing to say for it:

  65   │ 2020-02-11 21:26:54,854 Warning: remote clipboard request timed out
  66   │ 2020-02-11 21:26:54,854  request id 2, selection=CLIPBOARD, target=TARGETS
  67   │ 2020-02-11 21:26:58,114 client   1 @44.787 window-close event on desktop or shadow window, disconnecting
  68   │ 2020-02-11 21:27:00,836 xpra client 1 disconnected.

but I didn't close the window

#2586 fixed re-enable systemd-run by default Antoine Martin Antoine Martin
Description

Following ticket:2585#comment:1 : in "auto" mode we now test it before trying to use it, so it should be safe to enable everywhere.

See also: #1105, #2042, #1536.

#2585 fixed Systemd cgroups ticket back at it again Tmanok Tmanok
Description

Hello XPRA Devs,

You all make a most excellent application, I use it daily as of this past month and used to use it long ago. There was a bug that plagued XPRA for about 2 years it seemed and it related to systemd changes. This month I have noticed this problem on every fresh install (from your APT repository) on over 10 devices just this week alone. My platforms are Ubuntu, Debian, LinuxMint, (and HeliumBL actually). My installation process is simple:

  1. sudo apt update ; sudo apt install apt-transport-https
  2. wget -q https://xpra.org/gpg.asc -O- | sudo apt-key add -
  3. sudo add-apt-repository "deb https://xpra.org/ bionic main"
  4. sudo apt update ; sudo apt install xpra
  5. xpra --version

(Should be 3.0.5 or greater in 2020.)

  1. xpra start :100 --start=/usr/bin/firefox

But then this happens:

using systemd-run to wrap 'start' server command
'systemd-run' '--description' 'xpra-start' '--scope' '--user' 'python3' '/usr/bin/xpra' 'start' '--start=/usr/bin/firefox' '--systemd-run=no'
Job for run-rfff30b9e85394e8198e446d53091aa64.scope failed.
See "systemctl status run-rfff30b9e85394e8198e446d53091aa64.scope" and "journalctl -xe" for details.
  1. journalctl -xe
    -- Unit xpra.socket has finished starting up.
    -- 
    -- The start-up result is RESULT.
    Feb 10 13:56:36 MintyCraptop sudo[7753]: pam_unix(sudo:session): session closed for user root
    Feb 10 13:57:18 MintyCraptop dbus-daemon[646]: Unknown group "power" in message bus configuration file
    Feb 10 13:57:18 MintyCraptop dbus-daemon[646]: Unknown username "whoopsie" in message bus configuration file
    Feb 10 13:57:18 MintyCraptop dbus-daemon[646]: [system] Reloaded configuration
    Feb 10 13:57:19 MintyCraptop sudo[7497]: pam_unix(sudo:session): session closed for user root
    Feb 10 13:57:45 MintyCraptop polkitd(authority=local)[675]: Registered Authentication Agent for unix-process:8190:9116487 (system bus name :1.1871 [/usr/bin/pkttyagent --notify-fd 5 --fallback
    Feb 10 13:57:45 MintyCraptop systemd[1442]: run-rfff30b9e85394e8198e446d53091aa64.scope: Failed to add PIDs to scope's control group: Permission denied
    Feb 10 13:57:45 MintyCraptop systemd[1442]: run-rfff30b9e85394e8198e446d53091aa64.scope: Failed with result 'resources'.
    Feb 10 13:57:45 MintyCraptop systemd[1442]: Failed to start xpra-start.
    -- Subject: Unit UNIT has failed
    -- Defined-By: systemd
    -- Support: http://www.ubuntu.com/support
    -- 
    -- Unit UNIT has failed.
    

So how do I resolve it? First I make sure that firefox or whatever the initial program is, isn't running (though it would be nice if this didn't have to happen).

ps aux | grep firefox
technic+  8327  0.0  0.0  14428  1012 pts/1    S+   14:10   0:00 grep --color=auto firefox

If that doesn't prove to yield anything as shown above, I then borrow the following information I found in a previous systemd ticket on GitHub and apply the "fix", though this is most certainly not from my understanding a very good solution and should only be done to personal machine and not within production environments...

#!/bin/bash

set -euo pipefail

file=/sys/fs/cgroup/user.slice/user-$(id -u).slice/cgroup.procs

sudo chown $(whoami):root $file
sudo chmod g+w            $file

Thanks, let me know if there's a missing dependency or something you want me to test though I really feel like this is an XPRA bug as many of the systems I've tried this on were fresh installations and still encountered this bug.

1 2 3 4 5 6 7 8 9 10 11 12
Note: See TracQuery for help on using queries.