[X2Go-Dev] Bug#1259: Debian package from heuler repository fails to install

2018-02-19 Thread Martin
Package: python-x2go
Version: 0.5.99.0-0x2go1~git20180112.1367+.heuler.1

"apt-get install python-x2go" fails with

Setting up python-x2go (0.5.99.0-0x2go1~git20180112.1367+.heuler.1) ...
  File "/usr/lib/python2.7/dist-packages/x2go/tests/test_printing.py", line 46
print(_printing, file=tf)
 ^
SyntaxError: invalid syntax

dpkg: error processing package python-x2go (--configure):
 installed python-x2go package post-installation script subprocess returned 
error exit status 101
Errors were encountered while processing:
 python-x2go
E: Sub-process /usr/bin/dpkg returned an error code (1)

The bug was introduced by commit 327fa5e5f09d7fabb4d0052de1b2876c852797e1 (Port 
to Python3).
The attached patch restores compatibility with Python 2.6+.

Regards
Martin
--- test_printing.py.orig	2018-01-12 13:46:28.0 +0100
+++ test_printing.py	2018-02-19 17:07:44.433726565 +0100
@@ -17,6 +17,8 @@
 # Free Software Foundation, Inc.,
 # 51 Franklin St, Fifth Floor, Boston, MA 02110-1301, USA.
 
+from __future__ import print_function
+
 __package__ = 'x2go.tests'
 __name__= 'x2go.tests.test_printing'
 
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev

Re: [X2Go-Dev] [pulseaudio] main.c: User-configured server at localhost:, refusing to start/autospawn.

2018-02-19 Thread Mike DePaulo
Hi Robert,

Whether this is a bug in x2go or an issue with x2go interacts with
PulseAudio, please a file bug with our bug tracker.

I have some comments I would like to add, and I will mention them there.

Also, the high CPU usage may be a separate issue.

-Mike

On Fri, Feb 9, 2018 at 9:51 AM, Robert Kudyba  wrote:
> This seems to happen a couple of times a week. We are on Fedora 27. and just
> upgraded to
> x2goserver-4.1.0.0-0.0x2go1.0.git20180129.1511.heuler.fc27.x86_64
>
> Every so often the logs start filling up with these:
> Feb  9 09:28:33 erdos pulseaudio[2124]: [pulseaudio] main.c: User-configured
> server at localhost:41031, refusing to start/autospawn.
>
> Should we remind users to uncheck the audio box when logging in?
>
> I see there is bug 161, https://bugs.x2go.org/cgi-bin/bugreport.cgi?bug=161
> and 1143, https://bugs.x2go.org/cgi-bin/bugreport.cgi?bug=1143 that are
> quite old but perhaps related. here's a 'ps -auwx | grep x2go' when this
> happens:
>
> user1  20652  0.0  0.0 116572  1696 ?Ss   Feb06   0:00 /bin/bash
> /usr/bin/x2goruncommand 333 20569 user1-333-1517963407_stDMATE_dp32 50324
> mate-session nosnd D
>
> user1  20756  0.0  0.0  32340  1116 ?SFeb06   0:00
> /usr/bin/dbus-run-session /etc/x2go/Xsession
>
> user2 30305  0.0  0.0 116704  3232 ?Ss   Feb08   0:00 bash -l -c
> echo "X2GODATABEGIN:216fe65b-1b56-4b0f-99b9-4f5fd0da8929"; export
> PATH="/usr/local/bin:/usr/bin:/bin";export TERM="dumb"; x2goresume-session
> user2-334-1518126156_stDXFCE_dp32 fullscreen adsl 16m-jpeg-9 us query 0
> both; echo "X2GODATAEND:216fe65b-1b56-4b0f-99b9-4f5fd0da8929";
>
> user2 30396  0.0  0.0 116572  3184 ?SFeb08   0:29 /bin/bash
> /usr/bin/x2goresume-session user2-334-1518126156_stDXFCE_dp32 fullscreen
> adsl 16m-jpeg-9 us query 0 both
>
> user331106  0.0  0.0 116572  1692 ?Ss   Feb06   0:00 /bin/bash
> /usr/bin/x2goruncommand 331 30861 user3-331-1517954980_stDMATE_dp32 41031
> mate-session nosnd D
>
> user331230  0.0  0.0  32340  1116 ?SFeb06   0:00
> /usr/bin/dbus-run-session /etc/x2go/Xsession
>
> user2 32682  0.0  0.0 116704  3216 ?Ss   Feb08   0:00 bash -l -c
> echo "X2GODATABEGIN:c04d2709-19ac-4774-8a7e-0dee1c5164d2"; export
> PATH="/usr/local/bin:/usr/bin:/bin";export TERM="dumb"; x2goresume-session
> user2-335-1518126319_stDMATE_dp32 fullscreen adsl 16m-jpeg-9 us query 0
> both; echo "X2GODATAEND:c04d2709-19ac-4774-8a7e-0dee1c5164d2";
>
> user2 32731  0.0  0n.
> .0 116572  3136 ?SFeb08   0:28 /bin/bash
> /usr/bin/x2goresume-session user2-335-1518126319_stDMATE_dp32 fullscreen
> adsl 16m-jpeg-9 us query 0 both
>
> You can see the server load gets a bit high:
> top - 09:29:24 up 6 days, 16:37,  4 users,  load average: 2.22, 3.07, 2.84
> Tasks: 383 total,   2 running, 295 sleeping,   0 stopped,   0 zombie
> %Cpu(s):  0.8 us,  0.5 sy,  1.6 ni, 87.0 id,  9.8 wa,  0.1 hi,  0.2 si,  0.0
> st
> KiB Mem : 24670328 total,   502868 free,  2891660 used, 21275800 buff/cache
> KiB Swap: 40959996 total, 40601596 free,   358400 used. 21282084 avail Mem
>   PID USER  PR  NIVIRTRESSHR S  %CPU %MEM TIME+ COMMAND
>  3026  20   0  157764  12240   4772 R  41.2  0.0   0:00.07 x2gogetagentsta
>
> So I'm forced to kill the PIDs.
>
> After restarting x2goserver:
> Feb  9 09:32:37 erdos /usr/sbin/x2gocleansessions[4477]:
> user3-331-1517954980_stDMATE_dp32, pid 30861: cleaning up stale X11 socket
> file: /tmp/.X11-unix/X331
>
> Feb  9 09:32:37 erdos /usr/sbin/x2gocleansessions[4477]:
> user3-331-1517954980_stDMATE_dp32, pid 30861: cleaning up stale X11 lock
> file: /tmp/.X331-lock
>
> Side note, why does /var/log/dnf.log have these nightly?
> DEBUG Unknown configuration option: protect = 0 in
> /etc/yum.repos.d/x2go.repo
>
> ___
> x2go-dev mailing list
> x2go-dev@lists.x2go.org
> https://lists.x2go.org/listinfo/x2go-dev
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev

[X2Go-Dev] Bug#1258: Linux is affected

2018-02-19 Thread Mike DePaulo
I can confirm that Linux is affected (including on 4.1.1.0
specifically), and that the commit fixes it.
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev

[X2Go-Dev] session files in /tmp

2018-02-19 Thread yk+bug+x2go
On Linux, the X2Go server stores the session files in /tmp/.x2go-* .
Now, whenever a coworker leaves for vacation, those files are not
getting updated and thus get deleted by tmpwatch on CentOS 6 or
systemd-tmpfiles on CentOS 7; when the coworker comes back and tries to
resume the session, it goes into failed mode since those files are
gone, with no possibility of recovery.
The current workaround that I set up is to exclude /tmp/.x2go-* from
clean-up (e.g.[1]), but to prevent having to do it on every machine to
which we want to connect, it would be nice to move the folder to a
place that is not cleaned (at least as often). Although the Linux
Filesystem Hierarchy Standard (FHS) has been widely disputed, it seems
to be an admitted convention to put PID/lock/socket files and the like
in /run (or /var/run). Is there a specific reason why those files are
in /tmp and not in /var?

[1] On CentOS 7 for systemd-tmpfiles:
# cat /etc/tmpfiles.d/x2go.conf
#Type PathMode UID  GID  Age Argument
# Exclude X2Go session files
x /tmp/.x2go-*
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev

[X2Go-Dev] Bug#883: Bug #883 is back - this time, on Linux

2018-02-19 Thread Stefan Baur
Hi everyone,

I hate to say this, but ... bug #883 is back. Somehow.  And this time,
on Linux.

Situation:

I run a Debian-Jessie-based X2Go-ThinClientEdition, in Mate-MiniDesktop,
amd64 flavor.

Mate-MiniDesktop means that there is a local desktop, Firefox, xterm,
etc., and X2GoClient is set to run in Published Applications mode.

Whenever I try to start LibreOffice as a Published Application, it
pretty much instantly snaps to fullscreen and the window decoration is
gone, so there's no way to un-maximize or resize the window (Pressing
Alt+Space does bring up a menu, but it doesn't do anything).

It does not happen on Windows.  So I tried to resize the LibreOffice
window to a size way smaller than the screen while running windows, then
I closed the session, and started a new one via X2Go-TCE.  Sadly, even
though the window appears in the correct size for the fraction of a
second, it remains without window decoration, and instantly resizes to
full screen again. :'(

Starting on Windows, keeping the window in a less-than-fullscreen-sized
size, suspending the session, and resuming it on Linux works, though.

At first, I thought it might be a xinerama issue, but it also happens
with xinerama=false on Linux.

Any idea as to what could be causing this?

Kind Regards,
Stefan Baur

-- 
BAUR-ITCS UG (haftungsbeschränkt)
Geschäftsführer: Stefan Baur
Eichenäckerweg 10, 89081 Ulm | Registergericht Ulm, HRB 724364
Fon/Fax 0731 40 34 66-36/-35 | USt-IdNr.: DE268653243



signature.asc
Description: OpenPGP digital signature
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev

[X2Go-Dev] Bug#855: Update on file deletion issue in portable mode

2018-02-19 Thread Stefan Baur
Package: x2goclient
Version: 4.1.0.0
Severity: grave

When you install X2GoClient on Windows in portable mode, in the
directory some_dir\x2goclient, and create a batch file to start it in
porable mode as some_dir\x2goclient.cmd, with the following content:

@echo off
cd /d %~dp0\x2goclient
start "X2GoClient" x2goclient.exe --portable

Then, either upon connecting to a server (causing ssh\known_hosts to be
created/updated), or upon closing of x2goclient, the content of these
directories gets deleted (save for the file ssh\known_hosts):

some_dir\x2goclient\.ssh
some_dir\x2goclient\ssh

Obviously, this is NOT good, because people will be tempted to store
their private keys in that directory, only to find them gone afterwards.

My guess is that it hits .ssh as well because .ssh is no regular FAT
file name, but rather one that can only be created in cmd.exe with "md
.ssh", and probably only works on VFAT, not regular FAT, so a routine
looking to match ssh that isn't VFAT-safe will hit .ssh instead.

Could it be that we have a bug in x2goclient that purges the ssh
directory whenever known_hosts is accessed for writing, or similar?

Kind Regards,
Stefan Baur

-- 
BAUR-ITCS UG (haftungsbeschränkt)
Geschäftsführer: Stefan Baur
Eichenäckerweg 10, 89081 Ulm | Registergericht Ulm, HRB 724364
Fon/Fax 0731 40 34 66-36/-35 | USt-IdNr.: DE268653243



signature.asc
Description: OpenPGP digital signature
___
x2go-dev mailing list
x2go-dev@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-dev