• Gnome Desktop fails to load properly

    From Jim Beard@2:250/1 to All on Sat Aug 3 16:18:33 2019
    I boot to run level 3, and use startx to launch the Gnome desktop.

    This morning, I turned on the monitor, to see only the system monitoring messages usually visible when I hit Cntl-Alt-F12 and indeed Cntl-Alt-F1
    took me to the normal tty1 login screen.

    I logged in, but startx did not bring up the normal Mageia desktop screen,
    the top panels to MageiaWelcome (where you can click to minimize, maximize
    etc) did not appear, and an autostart program for multiple language
    characters had in its tool bar only the choice that would allow me to
    quit.

    Is there something to easily check and perhaps correct before I create a
    new user and see if that works?

    I have a full backup a few days old, but have no idea what or how much I
    would need to replace from backup.

    Before starting floundering around, does anyone have suggestions?

    Logging in to the machine using ssh from my backup machine (the one I am
    using at the moment) has crossed my mind, but I have never used it and
    would need instructions on how to set that up.

    Cheers!

    jim b.

    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Bit Twister@2:250/1 to All on Sat Aug 3 17:29:08 2019
    On Sat, 3 Aug 2019 15:18:33 -0000 (UTC), Jim Beard wrote:
    I boot to run level 3, and use startx to launch the Gnome desktop.

    This morning, I turned on the monitor, to see only the system monitoring messages usually visible when I hit Cntl-Alt-F12 and indeed Cntl-Alt-F1
    took me to the normal tty1 login screen.

    I logged in, but startx did not bring up the normal Mageia desktop screen, the top panels to MageiaWelcome (where you can click to minimize, maximize etc) did not appear, and an autostart program for multiple language characters had in its tool bar only the choice that would allow me to
    quit.

    Is there something to easily check and perhaps correct before I create a
    new user and see if that works?

    In my stupid opinion, you should always have a second user login for testing/proving if it is a user problem or system wide problem.

    I now create a "normal" account on clean install, then a "junk" account,
    then all my user accounts.

    "normal" is to have a known, pristine account as a reference.
    "junk" is used for using su to make all post install changes and for
    general testing there after.


    I have a full backup a few days old, but have no idea what or how much I would need to replace from backup.

    Yup, been there done that. I now have a user backup and a system backup
    script. User backup just does user files. System backup is nothing more
    than a rsync of / less the user's files to a /hotbu partition.

    Makes it dead easy to run restore_mga7 which rsync's /hotbu/ to mga7


    Before starting floundering around, does anyone have suggestions?

    Snippet derived from https://bugs.mageia.org/show_bug.cgi?id=16894
    startx does not start default desktop configured in /etc/sysconfig/desktop, defaults to IceWm

    Reboot runlevel 3
    get a user terminal and login
    startx startgnome


    Logging in to the machine using ssh from my backup machine (the one I am using at the moment) has crossed my mind, but I have never used it and
    would need instructions on how to set that up.

    Depends on your setup. I assign static ip addresses to all nodes on my LAN. Makes it simple to just do a ssh target_node_name_here.

    Should be easy enough to use systemctl to enable/start sshd on both machines, run "xhost xhost +$(hostname --fqdn)" then
    "ssh user@target_node_name_here" or "ssh user@target_node_ip_address_here"

    With sshd running on both systems you can ssh from/to either system.

    You can not ssh root@target. You use a user id and after logging in
    do a su - root.

    You can do a "ssh target" if you are going to use the same user on target node.
    example:
    $ ssh tb
    Warning: Permanently added 'tb' (ECDSA) to the list of known hosts.
    Last login: Sat Aug 3 02:02:47 2019 from wb.home.test

    $ whoami
    bittwister


    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Bit Twister@2:250/1 to All on Sat Aug 3 17:41:01 2019
    On Sat, 3 Aug 2019 11:29:08 -0500, Bit Twister wrote:
    On Sat, 3 Aug 2019 15:18:33 -0000 (UTC), Jim Beard wrote:

    Logging in to the machine using ssh from my backup machine (the one I am
    using at the moment) has crossed my mind, but I have never used it and
    would need instructions on how to set that up.

    Depends on your setup. I assign static ip addresses to all nodes on my LAN. Makes it simple to just do a ssh target_node_name_here.

    Should be easy enough to use systemctl to enable/start sshd on both
    machines,
    run "xhost xhost +$(hostname --fqdn)" then
    "ssh user@target_node_name_here" or "ssh user@target_node_ip_address_here"

    With sshd running on both systems you can ssh from/to either system.

    You can not ssh root@target. You use a user id and after logging in
    do a su - root.

    You can do a "ssh target" if you are going to use the same user on target
    node.
    example:
    $ ssh tb
    Warning: Permanently added 'tb' (ECDSA) to the list of known hosts.
    Last login: Sat Aug 3 02:02:47 2019 from wb.home.test

    $ whoami
    bittwister

    Frap, forgot to mention you may need to poke a hole in shorewall to
    allow ssh access. Example from /etc/shorewall/rules
    ACCEPT net:your_lan_ip_addresses_here all tcp ssh
    and "systemctl restart shorewall"

    your_lan_ip_addresses_here is a comma list of lan nodes,no spaces.
    example: 192.168.11.100,192.168.11.200

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From David W. Hodgins@2:250/1 to All on Sat Aug 3 20:15:05 2019
    On Sat, 03 Aug 2019 11:18:33 -0400, Jim Beard <jim.beard@verizon.net> wrote:

    I boot to run level 3, and use startx to launch the Gnome desktop.

    Mageia 6, 7, or cauldron?

    This morning, I turned on the monitor, to see only the system monitoring messages usually visible when I hit Cntl-Alt-F12 and indeed Cntl-Alt-F1
    took me to the normal tty1 login screen.

    That I haven't seen before, or heard about. Was the system left on with
    just the monitor turned off?

    I logged in, but startx did not bring up the normal Mageia desktop screen, the top panels to MageiaWelcome (where you can click to minimize, maximize etc) did not appear, and an autostart program for multiple language characters had in its tool bar only the choice that would allow me to
    quit.

    Is there something to easily check and perhaps correct before I create a
    new user and see if that works?

    I'd try the new user test first, as that's very easy to do and either
    narrows the problem down to a ~/someconfigfile, or rules that out.

    I have a full backup a few days old, but have no idea what or how much I would need to replace from backup.

    Before starting floundering around, does anyone have suggestions?

    Check /var/log/Xorg.0.log, "journalctl -b -1" (prior boot), and if present,
    the ~/.xsession-errors files for messages. If the error is present in the current boot, also check "dmesg|tail".

    Logging in to the machine using ssh from my backup machine (the one I am using at the moment) has crossed my mind, but I have never used it and
    would need instructions on how to set that up.

    That's only needed if you cannot access a working terminal using alt+ctrl+f2
    or similar.

    Regards, Dave Hodgins

    --
    Change dwhodgins@nomail.afraid.org to davidwhodgins@teksavvy.com for
    email replies.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Jim Beard@2:250/1 to All on Sat Aug 3 20:43:33 2019
    On Sat, 03 Aug 2019 15:15:05 -0400, David W. Hodgins wrote:

    On Sat, 03 Aug 2019 11:18:33 -0400, Jim Beard <jim.beard@verizon.net>
    wrote:

    I boot to run level 3, and use startx to launch the Gnome desktop.

    Mageia 6, 7, or cauldron?

    This morning, I turned on the monitor, to see only the system
    monitoring messages usually visible when I hit Cntl-Alt-F12 and indeed
    Cntl-Alt-F1 took me to the normal tty1 login screen.

    That I haven't seen before, or heard about. Was the system left on with
    just the monitor turned off?

    I am not sure, but think I was logged out.

    I logged in, but startx did not bring up the normal Mageia desktop
    screen,
    the top panels to MageiaWelcome (where you can click to minimize,
    maximize etc) did not appear, and an autostart program for multiple
    language characters had in its tool bar only the choice that would
    allow me to quit.

    Is there something to easily check and perhaps correct before I create
    a new user and see if that works?

    I'd try the new user test first, as that's very easy to do and either
    narrows the problem down to a ~/someconfigfile, or rules that out.

    The new test user worked.

    I have a full backup a few days old, but have no idea what or how much
    I would need to replace from backup.

    Before starting floundering around, does anyone have suggestions?

    Check /var/log/Xorg.0.log, "journalctl -b -1" (prior boot), and if
    present, the ~/.xsession-errors files for messages. If the error is
    present in the current boot, also check "dmesg|tail".

    I did not have an .xsession-errors file.

    Logging in to the machine using ssh from my backup machine (the one I
    am using at the moment) has crossed my mind, but I have never used it
    and would need instructions on how to set that up.

    That's only needed if you cannot access a working terminal using
    alt+ctrl+f2 or similar.

    I decided to go full-bore restore from backup, complicated by the latest version was some days old.

    As I could login to run level 3, I made a complete backup of the current $HOME, and then used rsync -av /backup/home/jim/ /home/jim .

    Not everything was as up to date as it should be so i did an rsync -av
    on /backup/home/jim/.config/evolution/ /home/jim/.config/evolution
    and that seemed to straighten things out.

    Then my attempt to run pan failed, so I edited the news servers
    preference files to match that of my backup machine, and pan is now
    working.

    I think I will use this machine as is for a while, make another complete backup of $HOME, and then use gnome-tweaks to try to get things back
    where they were yesterday. The install of Mageia 7 was followed by a
    copy in from Mageia 6 of $HOME so gnome-tweaks were in place, but my
    attempts to improve things may have tangled up something. All had been working though, so I was a bit surprised at my problem this morning.

    Cheers!

    jim b.





    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Jim Beard@2:250/1 to All on Sat Aug 3 20:49:22 2019
    On Sat, 03 Aug 2019 11:29:08 -0500, Bit Twister wrote:

    On Sat, 3 Aug 2019 15:18:33 -0000 (UTC), Jim Beard wrote:
    I boot to run level 3, and use startx to launch the Gnome desktop.

    This morning, I turned on the monitor, to see only the system
    monitoring messages usually visible when I hit Cntl-Alt-F12 and indeed
    Cntl-Alt-F1 took me to the normal tty1 login screen.

    I logged in, but startx did not bring up the normal Mageia desktop
    screen,
    the top panels to MageiaWelcome (where you can click to minimize,
    maximize etc) did not appear, and an autostart program for multiple
    language characters had in its tool bar only the choice that would
    allow me to quit.

    Is there something to easily check and perhaps correct before I create
    a new user and see if that works?

    In my stupid opinion, you should always have a second user login for testing/proving if it is a user problem or system wide problem.

    I now create a "normal" account on clean install, then a "junk" account,
    then all my user accounts.

    "normal" is to have a known, pristine account as a reference.
    "junk" is used for using su to make all post install changes and for
    general testing there after.


    I have a full backup a few days old, but have no idea what or how much
    I would need to replace from backup.

    Yup, been there done that. I now have a user backup and a system backup script. User backup just does user files. System backup is nothing more
    than a rsync of / less the user's files to a /hotbu partition.

    Makes it dead easy to run restore_mga7 which rsync's /hotbu/ to mga7


    Before starting floundering around, does anyone have suggestions?

    Snippet derived from https://bugs.mageia.org/show_bug.cgi?id=16894
    startx does not start default desktop configured in
    /etc/sysconfig/desktop, defaults to IceWm

    Reboot runlevel 3 get a user terminal and login startx startgnome


    Logging in to the machine using ssh from my backup machine (the one I
    am using at the moment) has crossed my mind, but I have never used it
    and would need instructions on how to set that up.

    Depends on your setup. I assign static ip addresses to all nodes on my
    LAN.
    Makes it simple to just do a ssh target_node_name_here.

    Should be easy enough to use systemctl to enable/start sshd on both
    machines,
    run "xhost xhost +$(hostname --fqdn)" then "ssh
    user@target_node_name_here" or "ssh user@target_node_ip_address_here"

    With sshd running on both systems you can ssh from/to either system.

    You can not ssh root@target. You use a user id and after logging in do a
    su - root.

    You can do a "ssh target" if you are going to use the same user on
    target node.
    example:
    $ ssh tb Warning: Permanently added 'tb' (ECDSA) to the list of known
    hosts.
    Last login: Sat Aug 3 02:02:47 2019 from wb.home.test

    I have copied the above, and added your comments on poking a hole in shorewall. For the moment, I will simply keep a couple of copies of that
    on each of my main and backup machines, for use if needed.

    Many thanks to you and Dave.

    Cheers!

    jim b.


    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Bit Twister@2:250/1 to All on Sat Aug 3 20:51:13 2019
    On Sat, 3 Aug 2019 19:43:33 -0000 (UTC), Jim Beard wrote:

    I think I will use this machine as is for a while, make another complete backup of $HOME, and then use gnome-tweaks to try to get things back
    where they were yesterday. The install of Mageia 7 was followed by a
    copy in from Mageia 6 of $HOME so gnome-tweaks were in place, but my
    attempts to improve things may have tangled up something. All had been working though, so I was a bit surprised at my problem this morning.

    Upon logout, mga6 gnome may have been dinked up by mga7 gnome tweaks.

    I can recommend getting ssh working between all nodes.
    Much easier to use gui tools rather than cli at runlevel3. :)

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Jim Beard@2:250/1 to All on Sat Aug 3 22:10:45 2019
    On Sat, 03 Aug 2019 14:51:13 -0500, Bit Twister wrote:

    On Sat, 3 Aug 2019 19:43:33 -0000 (UTC), Jim Beard wrote:

    I think I will use this machine as is for a while, make another
    complete backup of $HOME, and then use gnome-tweaks to try to get
    things back where they were yesterday. The install of Mageia 7 was
    followed by a copy in from Mageia 6 of $HOME so gnome-tweaks were in
    place, but my attempts to improve things may have tangled up something.
    All had been working though, so I was a bit surprised at my problem
    this morning.

    Upon logout, mga6 gnome may have been dinked up by mga7 gnome tweaks.

    I can recommend getting ssh working between all nodes.
    Much easier to use gui tools rather than cli at runlevel3. :)

    ssh is now working between my main and backup machines. I use nfs on my
    lan, and that seems to take care of shorewall requirements.

    The advantage of ssh is that I can use commands on the target machine,
    which is often critical.

    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh
    got me an error message that Only console users are allowed to use the X server.

    Cheers!

    jim b.


    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Bit Twister@2:250/1 to All on Sat Aug 3 22:55:13 2019
    On Sat, 3 Aug 2019 21:10:45 -0000 (UTC), Jim Beard wrote:


    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh
    got me an error message that Only console users are allowed to use the X server.

    Yep, let target system boot runlevel5/gui and you should have X server
    access.


    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Jim Beard@2:250/1 to All on Sun Aug 4 00:04:31 2019
    On Sat, 03 Aug 2019 16:55:13 -0500, Bit Twister wrote:

    On Sat, 3 Aug 2019 21:10:45 -0000 (UTC), Jim Beard wrote:


    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh got me
    an error message that Only console users are allowed to use the X
    server.

    Yep, let target system boot runlevel5/gui and you should have X server access.

    I rebooted my backup machine to run level 5, and
    ssh me@backup.home.invalid took me to a command line and the message
    that only those on console were allowed to use X.

    Either I am leaving out critical steps or parameters, or I need magical incantations to make it work.

    Cheers!

    jim b.

    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From David W. Hodgins@2:250/1 to All on Sun Aug 4 00:52:07 2019
    On Sat, 03 Aug 2019 17:10:45 -0400, Jim Beard <jim.beard@verizon.net> wrote:

    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh
    got me an error message that Only console users are allowed to use the X server.

    #!/bin/bash
    killall x0vncserver
    x0vncserver display=:0 -SecurityTypes=None &
    sleep 4
    vncviewer -compresslevel 9 localhost:0
    killall x0vncserver

    The above will give you access to a gui that's already running on the
    target machine, provided you are sshing to the user that is running
    the gui.

    My old script for starting a new gui login using ssh isn't working.
    I'll have to learn what's changed before I can respond with that, or
    someone else may have a currently working script. The last time I
    modified that script was in 2011.

    Regards, Dave Hodgins

    --
    Change dwhodgins@nomail.afraid.org to davidwhodgins@teksavvy.com for
    email replies.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Jim Beard@2:250/1 to All on Sun Aug 4 03:53:42 2019
    On Sat, 03 Aug 2019 19:52:07 -0400, David W. Hodgins wrote:

    On Sat, 03 Aug 2019 17:10:45 -0400, Jim Beard <jim.beard@verizon.net>
    wrote:

    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh got me
    an error message that Only console users are allowed to use the X
    server.

    #!/bin/bash
    killall x0vncserver
    x0vncserver display=:0
    -SecurityTypes=None &
    sleep 4
    vncviewer -compresslevel 9 localhost:0
    killall x0vncserver

    The above will give you access to a gui that's already running on the
    target machine, provided you are sshing to the user that is running the
    gui.

    My system knows nothing about x0vncserver or vncviewer.
    I assume this involves installing vnc, but which vnc packages do I need, installed where?

    Possibilities seem to be vnc-server-common, x11-vnc (my system seems to
    be using wayland, so would this work?), tigervnc-server-module, tigervnc- server, tigervnc, and a variety of lib64.*vnc packages, which seem needed
    only for specific applications.

    What I would like to do is to run my evolution mail program from the
    machine it is on or from my backup machine. I would ssh in from my
    backup machine and run evolution on the main machine. With gnome desktop running on both backup and main machine, I can open an x-term window on backup, ssh into my main machine using my username (the same on both machines), and run evolution, but rather than launch evolution and send
    the output to me on backup it wants to create a whole new evolution.

    This may be asking too much. Just being able to ssh between the two
    machines is a substantial improvement over using nfs.

    Cheers!

    jim b.

    --
    UNIX is not user-unfriendly, it merely expects users to be computer-
    friendly.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From Bit Twister@2:250/1 to All on Sun Aug 4 05:56:13 2019
    On Sat, 3 Aug 2019 23:04:31 -0000 (UTC), Jim Beard wrote:
    On Sat, 03 Aug 2019 16:55:13 -0500, Bit Twister wrote:

    On Sat, 3 Aug 2019 21:10:45 -0000 (UTC), Jim Beard wrote:


    I seem to remember there is a way to use ssh and get a gui from the
    target machine. My attempt to run startx when logged in via ssh got me
    an error message that Only console users are allowed to use the X
    server.

    Yep, let target system boot runlevel5/gui and you should have X server
    access.

    I rebooted my backup machine to run level 5, and
    ssh me@backup.home.invalid took me to a command line and the message
    that only those on console were allowed to use X.

    Hmmm, I hope .invalid is not hurting you. I would recommend using .test.
    $ cat /etc/hostname
    wb.home.test

    $ hostname
    wb.home.test

    I remember I had a problem with some app/service when I started out
    with .invalid. After switching to .test it felt better. It has been
    a decade or so since so I have forgotten which app.

    Note: anytime you change host name, you need to reboot so that all services/daemons pick up the new hostname.

    Read this whole reply before doing anything. Steps not in logical order.
    Change .invalid to .test as last resort.

    Either I am leaving out critical steps or parameters, or I need magical incantations to make it work.

    Ok, easiest thing to do is debug on your system, then debug on target node. once "ssh $USER@$(hostname)" works on source and target node, you then
    test between systems.

    $ hostname --ip-address better not return 127.0.0.1

    Do use "ping -c1 node_of_intrest_here" to verify you are using/getting
    valid ip addresses for other nodes.

    Troubleshooting tips. As root
    grep -i forward /etc/ssh/ssh_config
    # ForwardAgent no
    ForwardX11 yes
    ForwardX11Trusted yes

    systemctl restart sshd if you change the above file.

    Add your user login id to the systemd-journal, all nodes, log out/in
    and you will be able to view journal without having to be root.

    On all systems, as user;
    cd ~/.ssh
    chmod 700 .
    chmod 600 *


    I would expect
    ssh $USER@$(hostname)
    to ssh into current system as yourself,

    echo $DISPLAY should have something like
    $ echo $DISPLAY
    localhost:10.0

    then run some gui type app like xeyes, xterm, kwrite, gedit, ...
    or a gui editor to verify it likes it.

    If fails, you click up another terminal and run
    journalctl --no-hostname -fa --no-pager
    and try the ssh again.

    If journal does not show any fault messages you use
    ssh -vv $USER@$(hostname)
    to get more debugging information.

    Looking around in my setup, I created a xx__local.sh file in
    /etc/profile.d which eventually runs
    xhost +$(hostname --fqdn)

    then looking through the script used to log into a node I run
    xhost +$_target_host
    ssh $_target_host

    /etc/hosts has all my nodes with FQDN and alias. Snippet
    192.168.1.100 tb.home.test tb
    192.168.1.132 wb.home.test wb
    192.168.1.200 mtv.home.test mtv

    I assume you have not dinked with /etc/hosts.allow or hosts.deny.


    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)
  • From David W. Hodgins@2:250/1 to All on Sun Aug 4 10:28:40 2019
    On Sat, 03 Aug 2019 22:53:42 -0400, Jim Beard <jim.beard@verizon.net> wrote:

    On Sat, 03 Aug 2019 19:52:07 -0400, David W. Hodgins wrote:
    #!/bin/bash
    killall x0vncserver
    x0vncserver display=:0
    -SecurityTypes=None &
    sleep 4
    vncviewer -compresslevel 9 localhost:0
    killall x0vncserver
    The above will give you access to a gui that's already running on the
    target machine, provided you are sshing to the user that is running the
    gui.

    My system knows nothing about x0vncserver or vncviewer.
    I assume this involves installing vnc, but which vnc packages do I need, installed where?

    They're in the packages tigervnc and tigervnc-server. Install them on
    the system you are connecting to, and that's where you run the above
    script.

    $ ssh mine
    Last login: Sun Aug 4 04:45:54 2019 from 192.168.10.2
    [dave@hodgins ~]$ rpm -qa|grep vnc|sort
    gtk-vnc-i18n-0.6.0-3.mga6
    lib64gtk-vnc2.0_0-0.6.0-3.mga6
    lib64gvnc1.0_0-0.6.0-3.mga6
    lib64vncserver1-0.9.12-1.1.mga6
    tigervnc-1.8.0-1.mga6
    tigervnc-server-1.8.0-1.mga6
    vnc-server-common-1.0-7.mga6
    x11vnc-0.9.16-1.mga6

    The server mine is defined in my ~/.ssh/config on the first system as

    Host *
    ServerAliveInterval 120

    Host mine
    Hostname 192.168.10.101
    Port 22222
    User dave
    PreferredAuthentications publickey

    Possibilities seem to be vnc-server-common, x11-vnc (my system seems to
    be using wayland, so would this work?), tigervnc-server-module, tigervnc- server, tigervnc, and a variety of lib64.*vnc packages, which seem needed only for specific applications.

    I haven't tried it on a wayland system. I think it should work but can't
    say for sure.

    What I would like to do is to run my evolution mail program from the
    machine it is on or from my backup machine. I would ssh in from my
    backup machine and run evolution on the main machine. With gnome desktop running on both backup and main machine, I can open an x-term window on backup, ssh into my main machine using my username (the same on both machines), and run evolution, but rather than launch evolution and send
    the output to me on backup it wants to create a whole new evolution.

    That should work. Running evolution on the main machine, with it's output displayed on the backup system. That doesn't need vnc, just ssh with ... [root@x3 ssh]# grep -v -e '^#' -e '^$' /etc/ssh/ssh_config
    ForwardX11 yes
    ForwardX11Trusted yes
    StrictHostKeyChecking no

    A gui program such as evolution should be fine, though you will find
    the display a little slow to update. I tried watching netflix via
    google-chrome over the ssh connection, but while it works the video
    can't keep up, even when it's just within my lan. The encryption/decryption
    of ssh slows it down too much. It's stronger encryption than what's used
    for https, but that comes at a price in terms of speed.

    This may be asking too much. Just being able to ssh between the two
    machines is a substantial improvement over using nfs.

    I use sshfs to access files on my old system ...
    [dave@x3 ~]$ cat bin/mountmine
    #!/bin/bash
    if [[ $(grep "dave@mine:/" /proc/mounts) ]]; then
    echo "dave@mine:/ already mounted"
    else
    sshfs dave@mine:/ /var/mnt/old -o allow_other
    fi

    I just run that script when I want access to the files on that system.
    Makes it much easier to copy or compare files between the systems.

    Regards, Dave Hodgins

    --
    Change dwhodgins@nomail.afraid.org to davidwhodgins@teksavvy.com for
    email replies.

    --- MBSE BBS v1.0.7.12 (GNU/Linux-x86_64)
    * Origin: A noiseless patient Spider (2:250/1@fidonet)