[X2Go-User] Linux client connects, but not Windows

[John Stoffel]

Hi Guys,
I've been a long time user of x2go and I love it, but recently I've
run into an issue where I can connect to my session from a linux
desktop, but not from a Windows desktop.

Linux Server system:
---
- It's a VM running with 24gb of RAM and 4 CPUs, running on ESX 6.0
- xfce 4.12 is my session manager and window manager.  
- CentOS 7.7.1908 (kernel 3.10.0-1062.1.2.el7.x86_64)
- I've got the following packages installed:
> rpm -qa |grep x2go
x2goserver-4.1.0.3-3.el7.x86_64
perl-X2Go-Log-4.1.0.3-3.el7.noarch
x2goserver-xsession-4.1.0.3-3.el7.noarch
x2goserver-common-4.1.0.3-3.el7.noarch
x2goagent-4.1.0.3-3.el7.x86_64
perl-X2Go-Server-4.1.0.3-3.el7.noarch
perl-X2Go-Server-DB-4.1.0.3-3.el7.x86_64

- xdpyinfo of the working server session show:

name of display::51.0
version number:11.0
vendor string:The X.Org Foundation
vendor release number:7000
X.Org version: 7.0.0
maximum request size:  16777212 bytes
motion buffer size:  256
bitmap unit, bit order, padding:32, LSBFirst, 32
image byte order:LSBFirst
number of supported pixmap formats:7
supported pixmap formats:
 depth 1, bits_per_pixel 1, scanline_pad 32
 depth 4, bits_per_pixel 8, scanline_pad 32
 depth 8, bits_per_pixel 8, scanline_pad 32
 depth 15, bits_per_pixel 16, scanline_pad 32
 depth 16, bits_per_pixel 16, scanline_pad 32
 depth 24, bits_per_pixel 32, scanline_pad 32
 depth 32, bits_per_pixel 32, scanline_pad 32
keycode range:minimum 8, maximum 255
focus:  window 0x422, revert to
PointerRoot
number of extensions:23
   BIG-REQUESTS
   DAMAGE
   DOUBLE-BUFFER
DPMS
   GLX
   MIT-SCREEN-SAVER
  MIT-SHM
   RANDR
   RECORD
   RENDER
  SECURITY
   SGI-GLX
  SHAPE
  SYNC
  X-Resource
  XC-MISC
   XFIXES
   XFree86-Bigfont
   XINERAMA
  XInputExtension
  XKEYBOARD
   XTEST
   XVideo



Linux desktop:
---
Fedora 30 (kernel 5.4.21-100.fc30.x86_64) with a 1900x1200 display.
- I've got the following packages installed:
   $ rpm -qa | grep x2go
   x2goclient-4.1.2.2-1.fc30.x86_64
   x2goserver-xsession-4.1.0.3-3.fc30.noarch
   x2goagent-4.1.0.3-3.fc30.x86_64
   x2goserver-common-4.1.0.3-3.fc30.noarch
   x2goserver-4.1.0.3-3.fc30.x86_64

Thought I'm only using the client, not the server.   This pair works
just fine, esp since they're across the country from each other with
about 80-100ms of delay, but a decently big pipe.

Windows Desktop (fails to run session)
---

My failing client is a Windows 10 system, and I just recently
installed the latest x2go client to see if that would help.  It's
running:

  X2Go Client v. 4.1.2.2 (Qt - 4.8.6)

And it's got the same 1900x1200 display attached to the PC, but it
also fails when using just the laptop display, or another smaller
remote display at home.  Annoying for sure!


When I connect from the failing client, it just closes the session
after quickly opening and showing the windows for a second, then
dying.

what's the best way to debug this, and is there any better way to
provide better debugging for end users when a connection is dropped
like this?



The session.log file doesn't seem to give any useful info when I
connect, it makes the connection, then it suspends almost
immediately.  Sorry for the crappy cut'n'paste here.  


Info: Not using local device configuration changes.
nxagentReconnectFailedFonts: WARNING! Font server tunneling not retrieved.
nxagentReconnectSession: WARNING! Unable to retrieve all the fonts currently in 
use. Missing fonts have been replaced.
nxagentSighupHandler: WARNING! Ignoring the signal in state [SESSION_GOING_UP].
Info: keyboard file created: 
'/home/stoffj/.x2go/C-stoffj-51-1571665940_stDXFCE_dp24/keyboard'
Info: Keycode conversion auto-determined as off
Info: re-reading keystrokes configuration
Info: using keystrokes file '/home/stoffj/.nx/config/keystrokes.cfg'
Warning: ignoring keystroke 'viewport_scroll_up' (already in use by 
'viewport_move_up')
Warning: ignoring keystroke 'viewport_scroll_up' (already in use by 
'viewport_move_up')
Warning: ignoring keystroke 'viewport_scroll_right' (already in use by 
'viewport_move_right')
Warning: ignoring keystroke 'viewport_scroll_right' (already in use by 
'viewport_move_right')
Warning: ignoring keystroke 'viewport_scroll_down' (already in use by 
'viewport_move_down')
Warning: ignoring keystroke 'viewport_scroll_down' (already in use by 
'viewport_move_down')
Current known keystrokes:
  close_session Ctrl+Alt+t
switch_all_screens Ctrl+Alt+f
  fullscreen Ctrl+Shift+Alt+f
minimize Ctrl+Alt+m
  defer Ctrl+Alt+e
ignore Ctrl+Alt+BackSpace
  force_synchronization Ctrl+Alt+j
resize Ctrl+Alt+r
  viewport_move_left Ctrl+Shift+Alt+Left
viewport_move_left Ctrl+Shift+Alt+KP_Left
  viewport_mo

Re: [X2Go-User] Clipboard doesn't work?

[John Stoffel]

Ulrich> There a occasional problems with clipboard when using
Ulrich> Windows. This is most like a bug in the Vcxsrv we ship which
Ulrich> has not been identified yet. However, I never heard of the
Ulrich> clipboard not working at all. 

A common problem I see is that when I copy and paste from the Linux
side (Fedora 28-29) and then paste into Windows 10, I get chinese
characters.  Which makes me think it's a UTF8 mismatch somewhere.

Ulrich> You could try using the built-in Xserver of Mobaxterm and
Ulrich> check if the problems persist.

I'll see what I can do next week when I find the time.  
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] X2go Connection / compression Settings defunct? Documentation?

[John Stoffel]

> "Ulrich" == Ulrich Sibiller  writes:

Ulrich> On Mon, Apr 29, 2019 at 3:42 PM Tristan Miller
Ulrich>  wrote:
>> 
>> Dear Nigra,
>> 
>> On Fri, 26 Apr 2019 18:48:34 -0700, Nigra Truo
>>  wrote:
>> > I'm forced to work with some extreme bandwith restrictions and was
>> > looking into using more compression and/or the Modem / ISDN connection
>> > setting. (I'm down to old time modem speeds, 128 kbit or less)
>> > So far I could not find any documentation that shows what these really
>> > do. I normally use the ADSL one, which works. The Modem and ISDN
>> > setting seems broken (and has been for as long as I rememeber), they
>> > don't really reduce the bandwidth usage nor improve latency.
>> > When switching to Modem, the screen updates actually get much worse, I
>> > can watch the system update the screen in lines (not the whole screen
>> > is updated in one go), which looks like a bug.
>> > I have had success setting the image quality to a lower setting, the
>> > screen looks worse, which is a good indicator that it is working.
>> > I can't find any documentation on what the difference is between
>> > different compression methods. Can you point me in the right
>> > direction?
>> > If I can find the info, I would be up for updating the documentation
>> > so that others can find it better.

Ulrich> The code sets tile size and defer parameters depending on the link
Ulrich> speed, but you can also set these parameters on your own, see man
Ulrich> nxagent:

Ulrich>defer=
Ulrich>defer image updates (enabled for all connection types
Ulrich> except LAN), accepts values 0, 1 and 2

Ulrich>The default value can be set via the command line
Ulrich> (-defer). The value provided as nx/nx option is set when resuming a
Ulrich> session, thus it overrides the command line default.

Ulrich>tile=
Ulrich>set the tile size in pixels (x) for bitmap data
Ulrich> sent over the wire

Ulrich>The default value can be set via the command line
Ulrich> (-tile). The value provided as nx/nx option is set when resuming a
Ulrich> session, thus it overrides the command line default.

Ulrich> You can add the desired valued to the X2GO_NXOPTIONS= line in
Ulrich> /etc/x2go/x2goagent.options, e.g. "menu=1,defer=2,tile=128x128"

Ulrich>  * Each defer level adds the following rules to the previous ones:
Ulrich>  *
Ulrich>  * Level 0  Eager encoding.
Ulrich>  *
Ulrich>  * Level 1  No data is put or copied on pixmaps, marking them always
Ulrich>  *  as corrupted and synchronizing them on demand, i.e. when
Ulrich>  *  a copy area to a window is requested, the source is syn-
Ulrich>  *  chronized before copying it.
Ulrich>  *
Ulrich>  * Level 2  The put images over the windows are skipped marking the
Ulrich>  *  destination as corrupted. The same happens for copy area
Ulrich>  *  and composite operations, spreading the corrupted regions
Ulrich>  *  of involved drawables.

Ulrich> Here are the defaults as found in the code:

Ulrich> case LINK_TYPE_MODEM:
Ulrich>   deferLevel = 2;
Ulrich>   tileWidth  = 64;
Ulrich>   tileHeight = 64;
Ulrich> case LINK_TYPE_ISDN:
Ulrich>   deferLevel = 2;
Ulrich>   tileWidth  = 64;
Ulrich>   tileHeight = 64;
Ulrich> case LINK_TYPE_ADSL:
Ulrich>   deferLevel = 2;
Ulrich>   tileWidth  = 4096;
Ulrich>   tileHeight = 4096;
Ulrich> case LINK_TYPE_WAN:
Ulrich>   deferLevel = 1;
Ulrich>   tileWidth  = 4096;
Ulrich>   tileHeight = 4096;
Ulrich> case LINK_TYPE_NONE:
Ulrich> case LINK_TYPE_LAN:
Ulrich>   deferLevel = 0;
Ulrich>   tileWidth  = 4096;
Ulrich>   tileHeight = 4096;
Ulrich> default:
Ulrich>   deferLevel = 0;
Ulrich>   tileWidth  = 64;
Ulrich>   tileHeight = 64;

Ulrich> To my surprise I found another parameter today which is currently
Ulrich> undocumented: -irlimit . This will limit the amount of data that
Ulrich> is used to update drawables on the real display. It can only be passed
Ulrich> via command line. To play with it add it to
Ulrich> /etc/x2go/x2goagent.options.

>> I asked pretty much the same question here in 2017.  The only response
>> I got was from another user who, like you and me, was only able to make
>> a few observations by trial and error testing:
>> 
>> 
>> I agree that it would be nice to get a comprehensive overview of the
>> different settings.

Ulrich> Well, I have done some research tonight:

Ulrich> First it is worth noting that what we are talking about here is the
Ulrich> way NX will compress images travelling the wire. NX also (delta)
Ulrich> compresses the message format that X11 uses. This is unaffected by the
Ulrich> pack method setting. So if you only have an xterm handling some text
Ulrich> you probalby do not see much difference after changing the pack method
Ulrich> because there are no

Re: [X2Go-User] x2go screen refresh broken on Fedora 28 client

[John Stoffel]

>>>>> "Stefan" == Stefan Baur  writes:

Stefan> Am 16.01.19 um 16:39 schrieb John Stoffel:
>> What's the trick to force the x2goclient to start a NEW session on the
>> same host?  For some reason, it's just automatically resuming, and not
>> giving me the option to select between different sessions.

Stefan> If you have only one suspended session, and it has the same connection
Stefan> options as the one you requested during reconnect, the default is to
Stefan> auto-reconnect. The only exception to this rule is Published Application
Stefan> mode, where you will always see the tile with the various options.

Stefan> Your options to avoid automatic session resuming are:

Stefan> - run x2goclient with parameter --no-autoresume
Stefan> - Create a session with slightly different settings, and use that to
Stefan>   connect (say, windowed instead of fullscreen)
Stefan> - force-terminate the session on the server, using a regular SSH session
Stefan>   and the commands x2golistsessions (to find out the session ID) and
Stefan>   x2goterminate-session SESSIONIDHERE.  The session ID is the second
Stefan>   column in the pipe-separated output by x2golistsessions.

Stefan,
THanks for the quick and helpful reply, I appreciate it!  As for my
problem, I ended up killing off and restarting my session on the
CentOS system.  I didn't have the time to debug, but if it happens
again, I'll do the --no-autoresume and see if it's a per-session
problem, or a system level problem.

Looking at the debug logs for x2goclient, I didn't really see anything
major.  Nor did I see anything in the logs on the server side which
really helped.  But something was messed up with the mouse and screen
refreshes for sure.

John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] x2go screen refresh broken on Fedora 28 client

[John Stoffel]

Dammit, it looks like it's really a server problem, because using the
latest window client also shows the problem.  Dang... Time to start a
new session on the server and see how that works.

___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

[X2Go-User] x2go screen refresh broken on Fedora 28 client

[John Stoffel]

Hi guys,
Like a fool I upgraded my Fedora 28 workstation, and now my
connection to my x2go session running on a CentOS 7.6.x server is
acting really strangely.

It will connect and show the windows, but then the display starts
flashing and the screen stops refreshing windows unless I'm actually
using a specific window.  Then it will work.  For example, I'm typing
this email in an xterm on system, which I'm using to SSH home to write
this email.

I had to refresh the terminal before it would update, but now that I'm
typing, it's working just fine.  But the rest of the windows in my
x2go session are NOT refreshing, and the screen isn't updating
properly either.  If I move the mouse, the screen *tries* to update
things, but it's really screwey.  While typing this in an xterm, if I
move the mouse, I have to refresh my mail client (emacs & viewmail) by
hand to get the display working again.

And it's very hard to get screen captures or cut'n'paste to show the
info.  But here goes...

Client:  Fedora 28 x86_64 system.  Updated as of Jan 16, 2019, which
broke the setup.

x2goserver-common-4.1.0.3-1.fc28.noarch
x2goserver-xsession-4.1.0.3-1.fc28.noarch
x2goserver-4.1.0.3-1.fc28.x86_64
x2goclient-4.1.1.1-1.fc28.x86_64
x2goagent-4.1.0.3-1.fc28.x86_64


Server: Redhat CentOS 7.6.1810 (Core)

x2goserver-common-4.1.0.3-1.el7.noarch
perl-X2Go-Log-4.1.0.3-1.el7.noarch
x2goagent-4.1.0.3-1.el7.x86_64
x2goserver-4.1.0.3-1.el7.x86_64
perl-X2Go-Server-DB-4.1.0.3-1.el7.x86_64
perl-X2Go-Server-4.1.0.3-1.el7.noarch
x2goserver-xsession-4.1.0.3-1.el7.noarch


Session: XFCE


I'm loathe to reboot the server, because I've got work in various
windows I need to keep going.

What's the trick to force the x2goclient to start a NEW session on the
same host?  For some reason, it's just automatically resuming, and not
giving me the option to select between different sessions.

I strongly suspect it's a problem with Fedora, since the update on
there of some libraries. Here's the dnf log from that upgrade:

2019-01-15T19:06:35Z DEBUG --> Starting dependency resolution
2019-01-15T19:06:35Z DEBUG ---> Package kernel-core.x86_64 4.19.14-200.fc28 
will be installed
2019-01-15T19:06:35Z DEBUG ---> Package kernel.x86_64 4.19.14-200.fc28 will be 
installed
2019-01-15T19:06:35Z DEBUG ---> Package kernel-devel.x86_64 4.19.14-200.fc28 
will be installed
2019-01-15T19:06:35Z DEBUG ---> Package kernel-modules-extra.x86_64 
4.19.14-200.fc28 will be installed
2019-01-15T19:06:35Z DEBUG ---> Package kernel-modules.x86_64 4.19.14-200.fc28 
will be installed
2019-01-15T19:06:35Z DEBUG ---> Package efivar.x86_64 35-1.fc28 will be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package efivar.x86_64 37-1.fc28 will be an 
upgrade
2019-01-15T19:06:35Z DEBUG ---> Package efivar-libs.x86_64 35-1.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package efivar-libs.x86_64 37-1.fc28 will be an 
upgrade
2019-01-15T19:06:35Z DEBUG ---> Package gnutls.x86_64 3.6.4-1.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package gnutls.x86_64 3.6.5-2.fc28 will be an 
upgrade
2019-01-15T19:06:35Z DEBUG ---> Package ibus-typing-booster.noarch 2.4.0-1.fc28 
will be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package ibus-typing-booster.noarch 2.4.1-1.fc28 
will be an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package kernel-headers.x86_64 4.19.13-200.fc28 
will be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package kernel-headers.x86_64 4.19.14-200.fc28 
will be an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package kernel-tools-libs.x86_64 
4.19.10-200.fc28 will be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package kernel-tools-libs.x86_64 
4.19.14-200.fc28 will be an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package krb5-libs.x86_64 1.16.1-21.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package krb5-libs.x86_64 1.16.1-24.fc28 will be 
an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package libcdr.x86_64 0.1.4-4.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package libcdr.x86_64 0.1.5-1.fc28 will be an 
upgrade
2019-01-15T19:06:35Z DEBUG ---> Package libetonyek.x86_64 0.1.8-1.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package libetonyek.x86_64 0.1.9-1.fc28 will be 
an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package libqxp.x86_64 0.0.1-2.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package libqxp.x86_64 0.0.2-1.fc28 nwill be an 
upgrade
2019-01-15T19:06:35Z DEBUG ---> Package librados2.x86_64 1:12.2.9-1.fc28 will 
be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package librados2.x86_64 1:12.2.10-1.fc28 will 
be an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package libreport.x86_64 2.9.5-1.fc28 will be 
upgraded
2019-01-15T19:06:35Z DEBUG ---> Package libreport.x86_64 2.9.5-3.fc28 will be 
an upgrade
2019-01-15T19:06:35Z DEBUG ---> Package libreport-filesystem.x86_64 
2.9.5-1.fc28 will be upgraded
2019-01-15T19:06:35Z DEBUG ---> Package libreport-filesystem.x86_64 
2.9.5-3.fc28 will be an upgrade
2019-01-15T19:06:35

Re: [X2Go-User] Attention all users of Pale Moon and X2Go

[John Stoffel]

In my case, I'm using x2go to manage a desktop that's 80ms away from
me, which obviously sucks for movies and all those animated
ads... tweaking my settings from WAN to ADSL helped.

It would be *nice* if there was some guidance, or possibly even
auto-tuning to show howto get the best performance over high bandwidth
but laggy links like this.

Should I be tweaking my compression settings as well?

And yes, I'll be trying out Pale Moon as well on my setup to see how
well it works for me in my situation.

It might be that I just need to use the web browser locally, and the
rest of my stuff is remote.  With a full screen remote session, it's
not clear how I can make this work on a single screen desktop.

John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] execute xmodmap on resume

[John Stoffel]

> "richard" == richard lucassen  writes:

richard> Hello list,
richard> Not an x2go problem, but maybe someone might be able to guess what is
richard> happening here. I made a chmod 755 shell script:

richard> $ cat /usr/lib/x2go/extensions/post-resume.d/010-xmodmap
richard> #!/bin/dash
richard> test -f ~/.Xmodmap || exit 0
richard> /usr/bin/xmodmap -display $DISPLAY ~/.Xmodmap

richard> When resuming an x2go session, the script is executed, but x2go session
richard> starts but does not show up. The problem is that in one or another way,
richard> the xmodmap executable is waiting for something, but I have no idea
richard> what. Putting the xmodmap line to the background using an "&" makes the
richard> x2go session start, but the xmodmap mappings are not set. I still have
richard> to set the key mappings manually by invoking "xmodmap ~/.Xmodmap".

It sounds like when the script runs, the Xserver isn't quite ready,
and xmodmap is waiting for that.  Maybe you can start debugging by
putting:

   xmodmap -verbose -pm

before your call, to see what it prints out.  And make sure the
$DISPLAY is set correctly too!

You can also do an 'strace xmodmap ...' call in there, to see where
it's going and what it's calling to.

John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

[X2Go-User] Best Linux Distro for x2go server sessions as of April 2018?

[John Stoffel]

Guys,

So what's the best linux distro to run x2go on these days, in terms of
the server, the system you connect to?  Which would people recommend?

RHEL 7.x is out, as are probably a bunch of the newer Gnome 3
desktops.  Does anything running MATE work well?  Running Debian 9.3
with lxde seems to work pretty well, using client 4.1.1.1 to connect.

But at my $WORK, where we're mostly RHEL/CentOS based, it looks like
sticking with CentOS 6.9 is the best solution currently.   Which kinda
stinks.  Does Fedora work better or at least just as well when running
the Mate desktop?

John

___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] MxLinux

[John Stoffel]

> "Robert" == Robert Dinse  writes:

Robert>   Found the issue with MxLinux, had monkeysphere
Robert> installed, it causes perl to be called with the -T flag, and
Robert> this made having insecure directories such as /etc or '.' in
Robert> your $PATH problematic.

So how did you debug this?  Did you find the problem in the x2go logs?
Or somewhere else?  Please share your expertise, since it's a pain to
debug otherwise.
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

[X2Go-User] Best client/server pairings for RHEL 6 server and linux/windows clients?

[John Stoffel]

Guys,

I'm using x2go server on a CentOS 6.9 box, kept quite upto date, and I
have no problems connecting to it from both the desktop of that
system, and from home when I'm connecting in from a Linux desktop.  My
versions are:

Server (Linux CentOS 6.9):

  x2goclient-4.0.1.4-1.el6.x86_64
  x2goserver-4.0.1.20-1.el6.x86_64
  x2golxdebindings-debuginfo-1.0.2.4-1.1.x86_64
  x2goserver-xsession-4.0.1.20-1.el6.x86_64
  python-x2go-0.5.0.3-1.el6.noarch
  x2goagent-3.5.0.32-3.1.x86_64


Linux Client (Linux Mint 18.3):
  ii  nxproxy 2:3.5.0.33-0x2go1+git20180204.776+8.main.1   amd64NX proxy
  ii  x2go-keyring 2012.07.23~jessie~main~17~build1 all  
GnuPG keys
 of all X2Go developers and the X2Go archive
  ii  x2goclient  4.1.1.1-0x2go1+git20180302.1732+8.main.1 amd64
X2Go Client
 application (Qt4)


But when I try to use the latest client stuff from my Windows 7
system, I just get a connection which shows up for a second, then
disconnects and returns me to the connection screen.  I can't figure
out what I'm doing wrong.  But I'm loathe to upgrade my server just in
case I break a working setup.

In the meantime I've built a new CentOS-7.4 system, which is using the
same packages as my CentOS 6.9 server, but for some reason I just
can't reliably connect there either.

Been following the mailing list for a while, since this is really my
best option to be able to have my linux sessions travel around with
me.  But debugging has been a chore, since there are so many files and
places which need to be looked into.  And it's not clear WHY
connections fail.

And actually, does RHEL/CentOS 7 with a Gnome desktop even work?  Is
there an issue with OpenGL like there is with VNC desktops?

Thanks,
John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] can't get x2g0 to work with wsl

[John Stoffel]

Troy> Ok, I'm very confused.. I've installed just about every x2go item in apt 
search
Troy> and still no go.
Troy> I'm running wsl, fully updated and upgraded.

What is 'wsl'?  Don't assume that anyone knows your environment.
Please specify what you have in excrutiating detail, so people can
replicate the issue. 

Troy> I've tried numerous docs and steps to get it to work and it always says it
Troy> can't  timeout connecting to 192.168.1.137:

Are you connecting from the same host, or a seperate host?

What versions of the software are you using?  Post the details.

What client OS are you using?  What version of the client software are
you using?

Troy> Can someone PLEASE help me figure this out.

sure, once you post some more useful information.

Troy> if I try the browser plugin it says "invalid plugin"
Troy> here's the main thing I've been following, as you can see session.conf is 
not
Troy> in /etc like he said it would be.

Does it work if you don't use the browser plugin?


Step back and see if you can make it work locally only on the server.
Then try with a full client remotely, then try the browser based
version.

John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

Re: [X2Go-User] Can't resume session on RHEL6 system properly

[John Stoffel]

>>>>> "Ulrich" == Ulrich Sibiller  writes:

Ulrich> On Thu, Jul 13, 2017 at 4:12 PM, John Stoffel  wrote:
>> 
Ulrich> That kde Session type is irrelevant. Your Session is probably broken 
beyond
Ulrich> repair. You can try sending SIGHUP to the x2goagent process and see if 
that
Ulrich> helps.
>> 
>> And there's no x2goagent process... just an nxagent process.  Damn.  I
>> wonder what happened.  I really like x2go, but debugging it is horrible...

Ulrich> Then try nxagent. I was unsure about the final procress name anyway ;-)

I tried doing a kill -HUP, and it killed it all off.  Oh well...

Ulrich> On the Server you can find a Logfile in ~/.x2go somewhere. Maybe it 
contains
Ulrich> some Information about what happened.
>> 
>> Nothing obivous unfortunately.  Any suggestions on how to make it more
>> stable?

Ulrich> You could try the current nx-libs beta:
Ulrich> http://wiki.x2go.org/doku.php/doc:howto:nx-libs-betatesting, it
Ulrich> contains lots of bugfixes.

Ulrich> Loop: PANIC! The remote NX proxy closed the connection.
Ulrich> Loop: PANIC! Failure negotiating the session in stage '7'.
Ulrich> Loop: PANIC! Wrong version or invalid session authentication cookie.

Ulrich> Are you really use the same version of nx-libs on both sides?

Ulrich> What packages (+ version) do you have running on both sides?

See my initial email, I gave the package information on both sides.
It's worked in the past without problems, but lately it's been much
more flaky.

Maybe I need to upgrade to the latest release on RHEL6, but from what
I see on the web pages, there are problems with the later versions
with GNOME based desktops.

So it's looking more and more like something else will have to take
over for my needs.  Not sure what...
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

[X2Go-User] Can't resume session on RHEL6 system properly

[John Stoffel]

Guys,
I've got an x2go session where it looks like there's some disagreement
internally between what the state really is on the system.  I'm seeing
the following messages in the /var/log/messages file:

   Jul 12 22:51:58 x2go-server /usr/sbin/x2gocleansessions[3243]: 
john-50-1498570300_stDGNOME_dp24: session status S desynchronized with current 
status (R) and session suspend already tried unsuccessfully
   Jul 12 22:52:12 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:52:30 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:52:49 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:53:08 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:53:28 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:53:47 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully
   Jul 12 22:54:06 x2go-server /usr/bin/x2gosuspend-session: session with ID 
john-50-1498570300_stDGNOME_dp24 has been suspended successfully


Note the first line?  Where is says the session status is desynchronized?  How 
would I fix this?

The server is running CentOS 6.9 (Final) x86_64:

   # rpm -qa |grep x2go
   x2goclient-4.0.1.4-1.el6.x86_64
   x2goserver-4.0.1.20-1.el6.x86_64
   x2golxdebindings-debuginfo-1.0.2.4-1.1.x86_64
   x2goserver-xsession-4.0.1.20-1.el6.x86_64
   python-x2go-0.5.0.3-1.el6.noarch
   x2goagent-3.5.0.32-3.1.x86_64
   x2gognomebindings-2.0.2.1-2.1.x86_64


And my client in a Linux Mint 18.2 x86_64 system with:

> dpkg-query -l |grep x2go
ii  nxproxy  
2:3.5.0.32-0x2go1+git20170412.734+8.main.1 amd64NX proxy
ii  pyhoca-cli   
0.5.0.3-0x2go1+git20170412.224+8.main.1all  Command line X2Go 
client written in Python
ii  pyhoca-gui   
0.5.0.6-0x2go1+git20170413.1337+8.main.1   all  Graphical X2Go client 
written in (wx)Python
ii  python-x2go  0.5.0.4-1  
all  Python module providing X2Go client API
ii  x2go-keyring 
2012.07.23~jessie~main~17~build1   all  GnuPG keys of all X2Go 
developers and the X2Go archive
ii  x2goclient   
4.1.0.0-0x2go1+git20170413.1549+8.main.1   amd64X2Go Client application 
(Qt4)


Now if I go and sit at the x2go-server system, and use the x2goclient
to connect to the session, it's pretty much going to work, though
sometimes it still fails and I need to wipe my working session.  Which
sucks.

I also see, on the client system the following in the file:

   $ cat ~/.x2go/S-john-50-1498570300_stDGNOME_dp24/sessions
   Loop: WARNING! Unrecognized session type 'unix-kde-depth_24'. Assuming agent 
session.
   Loop: WARNING! Unrecognized session type 'unix-kde-depth_24'. Assuming agent 
session.
   Loop: PANIC! The remote NX proxy closed the connection.
   Loop: PANIC! Failure negotiating the session in stage '7'.
   Loop: PANIC! Wrong version or invalid session authentication cookie.


But I never use kde as my session, and it's configured to use GNOME
only.  I've never used KDE at all, except to run their great CD buring
tool k3db at points.

What other debug info can I provide?

Thanks,
John
___
x2go-user mailing list
x2go-user@lists.x2go.org
https://lists.x2go.org/listinfo/x2go-user

[X2Go-User] x2goclient Mate 17.04 can't connect to CentOS 6.8

[John Stoffel]

Hi guys,
I'm running a CentOS 6.8 x86_64 system using the following versions of
x2go:

   > rpm -qa |grep x2go
   x2goclient-4.0.1.4-1.el6.x86_64
   x2goserver-4.0.1.20-1.el6.x86_64
   x2golxdebindings-debuginfo-1.0.2.4-1.1.x86_64
   x2goserver-xsession-4.0.1.20-1.el6.x86_64
   python-x2go-0.5.0.3-1.el6.noarch
   x2goagent-3.5.0.32-3.1.x86_64
   x2gognomebindings-2.0.2.1-2.1.x86_64

My home system is a Linux Mint Serena 18.1 MATE x86_64 system.  It's
having problems connecting to the CentOS system.

So I use this CentOS system as both the x2go server, and as the
client, so I can work from either home or work with the same session.
But now from my home system (Linux Mint Serena 18.1 x86_64), I can't
get a running session, it just hangs.  Running x2goclient --debug I
get the following:

x2go-DEBUG-../src/sshprocess.cpp:199> Executing remote command via 
SshProcess object 3: "x2goresume-session stoffj-50-1490735175_stDGNOME_dp24 
1800x1100 wan 16m-jpeg-9 us auto 1 both"
x2go-DEBUG-../src/sshprocess.cpp:204> this=SshProcess(0x264ec60)  Running 
masterCon->addChannelConnection(this, ' "47d7a6d9-893d-45ed-b62c-4fe558416916" 
', ' "bash -l -c 'echo "X2GODATABEGIN:47d7a6d9-893d-45ed-b62c-4fe558416916"; 
export PATH="/usr/local/bin:/usr/bin:/bin"; export TERM="dumb"; 
x2goresume-session stoffj-50-1490735175_stDGNOME_dp24 1800x1100 w" ');
x2go-DEBUG-../src/sshmasterconnection.cpp:1325> Locking SSH channel 
connection MUTEX.
x2go-DEBUG-../src/sshmasterconnection.cpp:1327> Passing new channel 
conenction object to channelConnections.
x2go-DEBUG-../src/sshmasterconnection.cpp:1329> Unlocking SSH channel 
connection MUTEX.
x2go-DEBUG-../src/sshmasterconnection.cpp:1517> Creating new channel.

x2go-DEBUG-../src/sshmasterconnection.cpp:1521> New channel:0x7fc7f4012790

x2go-DEBUG-../src/sshmasterconnection.cpp:1552> Executing remote: "bash -l 
-c 'echo "X2GODATABEGIN:47d7a6d9-893d-45ed-b62c-4fe558416916"; export 
PATH="/usr/local/bin:/usr/bin:/bin"; export TERM="dumb"; x2goresume-session 
stoffj-50-1490735175_stDGNOME_dp24 1800x1100 wan 16m-jpeg-9 us auto 1 both; 
echo "X2GODATAEND:47d7a6d9-893d-45ed-b62c-4fe558416916";'"

x2go-DEBUG-../src/sshmasterconnection.cpp:1575> New exec channel created.


And it just seems to hang here.  On the server, I see the following in 
/var/log/messages:

Apr 25 18:09:16 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:16 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:16 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:22 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:27 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:35 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:35 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:35 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully
Apr 25 18:09:41 sekrit /usr/bin/x2gosuspend-session: session with ID 
stoffj-50-1490735175_stDGNOME_dp24 has been suspended successfully

And on the x2goserver system, in the session.log file I see:

[root@sekrit C-stoffj-50-1490735175_stDGNOME_dp24]# tail session.log
Session: Suspending session at 'Tue Apr 25 18:11:31 2017'.
Session: Session suspended at 'Tue Apr 25 18:11:31 2017'.
Session: Resuming session at 'Tue Apr 25 18:11:31 2017'.
Warning: Option file doesn't contain a port specification.
Info: Proxy running in server mode with pid '14478'.
Info: Waiting for connection from any host on port '4050'.
Info: Aborting the procedure due to signal '1'.
Session: Display failure detected at 'Tue Apr 25 18:11:37 2017'.
Session: Suspending session at 'Tue Apr 25 18:11:37 2017'.
Session: Session suspended at 'Tue Apr 25 18:11:37 2017'.


so I'm wondering what has happened here?  I'm trying to update my CentOS system 
to the latest packages, but it's probably not going to help since I don't see 
any x2go packages in the update.

Is there anything else I can do to help debug this?

Thanks,
John
___
x2go-user mailing list
x2go-user@lists.x2go.org
http://lists.x2go.org/listinfo/x2go-user

[X2Go-User] Keyboard problems: Ctrl-Alt-Left works, but not Ctrl-Alt-Right

[John Stoffel]

Hi all,

I'm running x2go on CentOS 7.x using the following packages:

  x2goclient-4.0.1.4-1.el6.x86_64
  x2goserver-4.0.1.20-1.el6.x86_64

And when I run the server on my desktop, and then the client so I can
connect to the screen I've setup, I have some minor keyboard
problems.  I can use Ctrl-Alt-Left to move inside the Full Screen x2go
session to another panel.

But I cannot move to the right at all, nor does running xev inside the
x2go session show me hitting those keys.  Just to note, I'm not
running x2go and shadowing my :0 display (which might be better?)
because I want to lock my session when I leave the office, and pick it
up at home without worrying about people seeing my display while I'm
away.

In any case, my setxkbmap INSIDE my x2go session shows:

   > setxkbmap -query
   rules:  base
   model:  pc104
   layout: us
   options:ctrl:nocaps


While outside it shows:

  > setxkbmap -query
  rules:  evdev
  model:  pc104
  layout: us
  options:ctrl:nocaps

I'm actually using a tenkeyless CODE keyboard, but I don't think that
makes much difference here.  Using the 'gnome-keybinding-properties' I
have the "Switch to workspace on the (left|right) of the current
workspace" setup to be Ctrl+Alt+Left or Ctrl+Alt+Right from what I
see.

Using 'xev' both inside and outside of the x2go client I see different
keycodes for my arrow keys.  Outside, in the standard CentOS 7.x Gnome
desktop, I see keycoded 113 (left) and 114 (right), but inside x2go I
see 100 (left) and 102 (right) so there's some mapping going on
somewhere which isn't getting through properly.

For example, inside X2go session, when I press and release the left
arrow key, I get:

KeyPress event, serial 37, synthetic NO, window 0x741,
root 0x6e, subw 0x0, time 3164346691, (107,75), root:(109,832),
state 0x0, keycode 100 (keysym 0xff51, Left), same_screen YES,
XLookupString gives 0 bytes:
XmbLookupString gives 0 bytes:
XFilterEvent returns: False

KeyRelease event, serial 37, synthetic NO, window 0x741,
root 0x6e, subw 0x0, time 3164346811, (107,75), root:(109,832),
state 0x0, keycode 100 (keysym 0xff51, Left), same_screen YES,
XLookupString gives 0 bytes:
XFilterEvent returns: False


So I'm wondering a couple of things.

1. If I edit my ~/.x2go/config/config/keystrokes.cfg file, does it get
   re-read automatically or do I have to stop and restart the x2goclient
   or the entire x2goserver session to have my changes take effect?

2. Has anyone else seen this issue and have suggestions on how I can
   maybe get around this?

Thanks,
John

P.S.  I do also have various nx* packages installed, would these be
interferring?

> rpm -qa | grep nx | sort
freenx-0.7.3-9.4.el6.centos.x86_64
libNX_X11-3.5.0.32-3.el6.x86_64
libNX_Xcomposite-3.5.0.32-3.el6.x86_64
libNX_Xdamage-3.5.0.32-3.el6.x86_64
libNX_Xdmcp-3.5.0.32-3.el6.x86_64
libNX_Xext-3.5.0.32-3.el6.x86_64
libNX_Xfixes-3.5.0.32-3.el6.x86_64
libNX_Xinerama-3.5.0.32-3.el6.x86_64
libNX_Xpm-3.5.0.32-3.el6.x86_64
libNX_Xrandr-3.5.0.32-3.el6.x86_64
libNX_Xrender-3.5.0.32-3.el6.x86_64
libNX_Xtst-3.5.0.32-3.el6.x86_64
nxagent-3.5.0.32-3.el6.x86_64
nxclient-3.5.0-7.x86_64
nx-libs-3.5.0.32-3.el6.x86_64
nxproxy-3.5.0.32-3.el6.x86_64


Thanks!
John
___
x2go-user mailing list
x2go-user@lists.x2go.org
http://lists.x2go.org/listinfo/x2go-user