x2go-dev

x2go-dev@lists.x2go.org

1 participants
5399 discussions

Bug#1094: When connected to a x2go server that has nVidia's binary drivers most software will segfault.
by Matthew Moskewicz 11 May '17

11 May '17

i ran into this issue on ubuntu 16.04 (64 bit), and i can confirm that both workarounds (using Nouveau and setting LD_LIBRARY_PATH=/usr/lib/nvidia-375 ) fix the issue. to add a bit more info, it seems that in the failing case, glxinfo (or presumably the other apps), can't find the GLX_indirect library that it trys to load dynamically, which has something to glvnd: https://github.com/NVIDIA/libglvnd looking at ldd, you can see that glxinfo is finding the GL libraries fine, using the system-set ld.so.conf settings that the nvidia driver sets up: moskewcz@mari:/etc/ld.so.conf.d$ ldd /usr/bin/glxinfo linux-vdso.so.1 => (0x00007ffe4779c000) libGL.so.1 => /usr/lib/nvidia-375/libGL.so.1 (0x00007fcd87f35000) libX11.so.6 => /usr/lib/x86_64-linux-gnu/libX11.so.6 (0x00007fcd87bfb000) libc.so.6 => /lib/x86_64-linux-gnu/libc.so.6 (0x00007fcd87831000) libdl.so.2 => /lib/x86_64-linux-gnu/libdl.so.2 (0x00007fcd8762d000) libGLX.so.0 => /usr/lib/nvidia-375/libGLX.so.0 (0x00007fcd873fd000) ... the paths are set here: moskewcz@mari:/etc/ld.so.conf.d$ cat /etc/ld.so.conf.d/x86_64-linux-gnu_EGL.conf /usr/lib/nvidia-375 /usr/lib32/nvidia-375 using strace, we can see the difference that setting LD_LIBRARY_PATH makes: # working case, opens the correct lib: moskewcz@mari:/etc/ld.so.conf.d$ LD_LIBRARY_PATH=/usr/lib/nvidia-375 strace glxinfo 2>&1 | grep GLX_ind open("/usr/lib/nvidia-375/tls/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/nvidia-375/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = 4 # non-working case, fails to open libGLX_indirect.so: moskewcz@mari:/etc/ld.so.conf.d$ strace glxinfo 2>&1 | grep GLX_ind open("/lib/x86_64-linux-gnu/tls/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/x86_64-linux-gnu/tls/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/x86_64-linux-gnu/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/x86_64-linux-gnu/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/x86_64-linux-gnu/tls/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/x86_64-linux-gnu/tls/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/x86_64-linux-gnu/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/x86_64-linux-gnu/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/tls/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/tls/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/lib/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/tls/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/tls/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) open("/usr/lib/x86_64/libGLX_indirect.so.0", O_RDONLY|O_CLOEXEC) = -1 ENOENT (No such file or directory) [...snip many more repeats of failed-to-load libGLX_indirect...] however, it's not clear why, under x2go, the nvidia code isn't searching in /usr/lib/nvidia-375 -- but it's not clear what code it uses for the DL loading, maybe it is somewhere in the glvnd code. mwm

1 0

Bug#592: X2Go issue (in src:x2goclient) has been marked as pending for release
by Oleksandr Shneyder 10 May '17

10 May '17

tag #592 pending fixed #592 4.1.0.1 thanks Hello, X2Go issue #592 (src:x2goclient) reported by you has been fixed in X2Go Git. You can see the changelog below, and you can check the diff of the fix at: http://code.x2go.org/gitweb?p=x2goclient.git;a=commitdiff;h=68bbf32 The issue will most likely be fixed in src:x2goclient (4.1.0.1). light+love X2Go Git Admin (on behalf of the sender of this mail) --- commit 68bbf328132125eaad5c53b0ac82490bf818e42e Author: Oleksandr Shneyder <o.shneyder(a)phoca-gmbh.de> Date: Wed May 10 15:22:11 2017 +0200 Interaction with SSH server (for example for changing expired password). Fixes: #592. diff --git a/debian/changelog b/debian/changelog index 57e48d8..40e262f 100644 --- a/debian/changelog +++ b/debian/changelog @@ -155,6 +155,8 @@ x2goclient (4.1.0.1-0x2go1) UNRELEASED; urgency=medium - Disable sound button on direct RDP and XDMCP sessions. Set for direct XDMCP session autologin=true. Set for direct XDMCP session username=XDM. + - Interaction with SSH server (for example for changing + expired password). Fixes: #592. [ Robert Parts ] * New upstream version (4.1.0.1):

2 1

Bug#1172: Bug#1172: Bug#1172: ThunderBird crashes in PubApp mode on amd64
by Walid MOGHRABI 10 May '17

10 May '17

> It is not "unusable". But nx offers an old version of GLX only. And > there seem to be some applications that don't care for the version the > X server (nxagent in this case) reports. So what should we do? You're right, I'm probably a bit to0 rude with this but it's far too old and most applications just test if GLX is enabled without testing it's version or correct support for each extensions which leads to crashes at times. Since there is no gain in keeping it "as is" (at least, I did see more problems than gains) and until GLX support is updated through Lib-NX, I personnaly would recommend disabling GLX by default (which let the possibility to the user to activate it if he really needs it but don't make this the default behaviour). Since applications are checking if GLX is enabled, disabling it force them to fallback on software renderers only (or don't run at all but I prefer that to crashes for unknown reasons). > It is possible to implement a newer GLX version but that would need > some investing quite some time. Unfortunately the current nx > development staff does not have that time... I know and I upvote for that as top priority since more and more applications needs this. Regards, Walid Moghrabi TRAVAUX.COM BAT I - PARC CEZANNE 2 290 AVENUE GALILEE - CS 80403 13591 AIX EN PROVENCE CEDEX 3 ----- Mail original ----- De: "Ulrich Sibiller" <uli42(a)gmx.de> À: "Walid MOGHRABI" <w.moghrabi(a)servicemagic.eu>, 1172(a)bugs.x2go.org Envoyé: Vendredi 5 Mai 2017 21:41:47 Objet: Re: [X2Go-Dev] Bug#1172: Bug#1172: ThunderBird crashes in PubApp mode on amd64 On Fri, May 5, 2017 at 5:33 PM, Walid MOGHRABI <w.moghrabi(a)servicemagic.eu> wrote: > I would recommend disabling GLX by uncommenting this line in /etc/x2go/x2goagent.options : > > #X2GO_NXAGENT_DEFAULT_OPTIONS+=" -extension GLX" > > GLX is unusable with the current NX-Libs (even Arctica ones) but since the extension is enabled, some components still tries to use H/W acceleration for some parts. > I get far better results (especially stability wise) with this option disabled. > > I would even consider disabling it "by default" until a proper GLX support is added to nx-libs. It is not "unusable". But nx offers an old version of GLX only. And there seem to be some applications that don't care for the version the X server (nxagent in this case) reports. So what should we do? It is possible to implement a newer GLX version but that would need some investing quite some time. Unfortunately the current nx development staff does not have that time... Regarding Thunderbird: does it still crash after setting gfx.xrender.enabled to true in about:config (respectively user.js)? Uli --- DISCLAIMER: This e-mail is private and confidential and may contain proprietary or legally privileged information. It is for the intended recipient only. If you have received this email in error, please notify the author by replying to it and then destroy it. If you are not the intended recipient you must not use, disclose, distribute, copy, print or rely on this e-mail or any attachment. Thank you

1 0

Bug#1172: ThunderBird crashes in PubApp mode on amd64
by Stefan Baur 09 May '17

09 May '17

package: nx-libs Hi, since Debian Jessie (stable) switched from IceDove back to ThunderBird, TB crashes when the host is an amd64 machine and X2GoClient is running in PubApp mode. i386 is unaffected. It happens with X2GoClient on Windows as well as on Linux, so it's not a matter of VcXsrv being too old/broken/whatever. It also happens when all X2Go server components are installed from heuler. It does NOT happen when you build nx-libs from source using the Debian Experimental repository. So there's something in the stable nx-libs that's broken on amd64 only, and got fixed in the nx-libs version that is available via Debian Experimental. Core-Devs, please advise how to fix this - should we dare making a new nx-libs release and declare it stable, based on what is in Debian Experimental (arguing that everything is better than what we currently have)? Or should we try to hunt down the offending code part and only patch that? If the latter, who's willing to take a stab at it? (We don't have a business case ATM, as all my TB-PubApp-users are still on i386, so no funding as of now.) 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

4 15

Bug#1016: X2Go issue (in src:python-x2go) has been marked as pending for release
by Mike Gabriel 09 May '17

09 May '17

tag #1016 pending fixed #1016 0.5.0.5 thanks Hello, X2Go issue #1016 (src:python-x2go) reported by you has been fixed in X2Go Git. You can see the changelog below, and you can check the diff of the fix at: http://code.x2go.org/gitweb?p=python-x2go.git;a=commitdiff;h=3ed4fed The issue will most likely be fixed in src:python-x2go (0.5.0.5). light+love X2Go Git Admin (on behalf of the sender of this mail) --- commit 3ed4fed3465e02c605c80e057c265ca4a5d4b2ac Author: Mike Gabriel <mike.gabriel(a)das-netzwerkteam.de> Date: Tue May 9 12:24:20 2017 +0200 Don't blindly release gevent locked. We need to checked if a semaphore is locked in some case and only then release it. (Fixes: #1016). diff --git a/debian/changelog b/debian/changelog index aeca116..cd549f0 100644 --- a/debian/changelog +++ b/debian/changelog @@ -3,6 +3,9 @@ python-x2go (0.5.0.5-0x2go1) UNRELEASED; urgency=low [ Mike Gabriel ] * New upstream version (0.5.0.5): - documentation: Fix wording in docstring. + - Don't blindly release gevent locked. We need to checked if a + semaphore is locked in some case and only then release it. + (Fixes: #1016). [ Mihai Moldovan ] * New upstream version (0.5.0.5):

2 1

Bug#1170: x2goplugin / x2goplugin-provider Failed to connect: Permission denied
by Justin St. Marie 08 May '17

08 May '17

Package: x2goserver Version: 4.0.1.20-1.el7.x86_64 Package: x2goplugin-provider Version: 4.1.0.0-1.el7.x86_64 Package: x2goplugin Version: 4.1.0.0-1.el7.x86_64 All packages are from the official EPEL 7 repository. When trying to log in using the firefox plugin, the error is: Failed to connect: Permission denied. The same credentials work with the x2go client. I used the example X2Go Demo config file, only changing server=IPaddress.

3 4

Bug#1172: Error message shown on tty1 when session crashes
by Stefan Baur 07 May '17

07 May '17

traps: x2goagent [3339] general protection ip:7f02464e2e6d sp:7ffef6c22150error:0 in libNX_X11.so.6.2[7f0246496000+123000] -Stefan -- 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

1 0

Processed: your mail
by owner＠bugs.x2go.org 06 May '17

06 May '17

Processing commands for control(a)bugs.x2go.org: > reassign 1173 x2goserver Bug #1173 [x2go-server] stDXFCE_dp24: is blocked Warning: Unknown package 'x2go-server' Bug reassigned from package 'x2go-server' to 'x2goserver'. Ignoring request to alter found versions of bug #1173 to the same values previously set Ignoring request to alter fixed versions of bug #1173 to the same values previously set > reassign 1174 x2goserver Bug #1174 [x2go-server] file is blocked (stDXFCE_dp24) Warning: Unknown package 'x2go-server' Bug reassigned from package 'x2go-server' to 'x2goserver'. No longer marked as found in versions 4.0.1.20. Ignoring request to alter fixed versions of bug #1174 to the same values previously set > merge 1173 1174 Bug #1173 [x2goserver] stDXFCE_dp24: is blocked Bug #1174 [x2goserver] file is blocked (stDXFCE_dp24) Merged 1173 1174 > End of message, stopping processing here. Please contact me if you need assistance. -- 1173: http://bugs.x2go.org/cgi-bin/bugreport.cgi?bug=1173 1174: http://bugs.x2go.org/cgi-bin/bugreport.cgi?bug=1174 X2Go Bug Tracking System Contact owner(a)bugs.x2go.org with problems

1 0

Bug#1174: file is blocked (stDXFCE_dp24)
by Felix B. Müller 05 May '17

05 May '17

Package: x2go-server Version: 4.0.1.20 Dear all, I have an Ubuntu 16.04.2 LTS (Server) which is integrated in the Windows Active Directory (AD). The user authentification is done via Kerberos followed here: https://help.ubuntu.com/community/ActiveDirectoryWinbindHowto The directories are also mounted automatically every time the user logs on via pam_mount: <volume user="*" fstype="cifs" server="fileserver1" path="home/%(DOMAIN_USER)" mountpoint="/home/%(DOMAIN_USER)"/> This works quite nice and I can login via ssh with the user names of the AD. I also would like to use x2go for the AD users. However, it works fine when I try to connect to the server (called ssh or ssh2) from the internal network (192.168.0.0). The users get their remote desktops. However, if they try to login remotely from another subnet the session could not be initialized. I was looking for the problem the last 3 weeks. Of course the first idea was due to a misconfigured firewall. The configuration is as follows: INTERNET ----- NAT1 -----(192.168.183.0)------ NAT2 ----(192.168.0.0)------ ssh-server I replaced the firewall, I put the PC (ssh-server) directly after the NAT. I had a look in all network connections with wireshark. I opened all ports. None of these things solved the problem. However, I can login with a local user (user account on the ssh-server) on the ssh-server but not with a user of the AD (username test01). I set the logging to the debug mode and I see that a file is always blocked. I do not understand why this file is blocked if I login from another subnet. Maybe this is not a bug I do not know and I do not understand this behavior. May 5 16:08:41 ssh2 /usr/sbin/x2gocleansessions[1717]: test01-50-1493993318_stDXFCE_dp24: updating session status from 'R' to 'F'. May 5 16:08:41 ssh2 /usr/sbin/x2gocleansessions[1717]: test01-50-1493993318_stDXFCE_dp24: is blocked. May 5 16:08:41 ssh2 /usr/sbin/x2gocleansessions[1717]: test01-50-1493993318_stDXFCE_dp24: adding to finished list. May 5 16:08:41 ssh2 /usr/bin/x2goumount-session[11035]: x2goumount-session has been called with options: Any ideas? Best regards!

1 0

Bug#1173: stDXFCE_dp24: is blocked
by Felix B. Müller 05 May '17

05 May '17

1 0

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

x2go-dev