GoogleEarth-5.2 install to Tumbleweed no longer possible?

It was still possible on oss-13.2

I try to install ge-5.2 on Tumbleweed following this link:

http://trixtar.org/mm/googleearth-5.2-linux-install.png

but get some missing files complaint (below).

until I overcome this I can’t really use Tumbleweed

It was bad enough not to be able to use any recent version of GE because past 5.2 there is no stick support, now it looks like the last possible workaround has been blocked too.

In the installedr crash log:

Major Version 5
Minor Version 2
Build Number 0001
Build Date Sep 1 2010
Build Time 11:25:42
OS Type 3
OS Major Version 4
OS Minor Version 10
OS Build Version 8
OS Patch Version 0
Crash Signal 6
Crash Time 1492141853
Up Time 0.877411

Stacktrace from glibc:
./libgoogleearth_free.so(+0xd090b)[0xf76f890b]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf7756e50]
[0x0]

In the cLi window:

Gtk-WARNING **: Failed to load module “libcanberra-gtk-module.so”: libcanberra-gtk-module.so: cannot open shared object file: No such file or directory
loki_setup: Suspect size value for option option

loki_setup: Suspect size value for option option

Installing mimetypes…
Installing desktop menu entries…
Installing desktop icon…
libGL error: unable to load driver: nouveau_dri.so
libGL error: driver pointer missing
libGL error: failed to load driver: nouveau
Fatal error in __driConfigOptions line 1, column 0: out of memory.
Google Earth has caught signal 6.

We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/root/.googleearth/crashlogs/crashlog-58f048a1.txt

Please include this file if you submit a bug report to Google.

Please don’t take this a dis but you are aware the latest stable Google Earth is
Earth version 7.1.8.3036

Thanks for taking the time to respond, I went to the trouble of downloading and installing it, alas NOTHING in this respect seems to have changed since version 5.2 after which GE went to LSB (Linux Standard Base) which it would seem STILL lacks control stick support fpr some reason. Starting the Flt-Sim the Control-Stick box in the launch dialogue is dead/ghosted. The stick is seen by oss-13.2/TW, I can even calibrate it but NOT in GE because as I understand it stick-suppert it is STILL not in LSB. I even tried to fly it with the keyboard which by definition would be nonsense but even there as an example the arrow keys work backwards. This is fine if you try to fly inverted for the first time without knowing that ailerons would/should then work inverted so that essentially it’s still a fubar :slight_smile:

I’m 73 and a retired pilot, the Ge-Flt-Sim is my only use for a computer or for the internet so without it it’s gonna be a short affair by the looks of things. It’ll work on 13.2 but eventually 13.2 won’t be usable on the net as generally every system becomes obsolete/unusable in no time.

I just installed to Debian 8 without any probs, so what’s up?

you can get libcanberra-gtk-module.so from the libcanberra-gtk2-module package from the Gnome:Factory repo
personally I don’t recommend having too many repo’s enabled but there was a recent change in TW which introduced vendor stickyness so adding Gnome:Factory is not a bad solution

zypper ar -f https://download.opensuse.org/repositories/GNOME:/Factory/openSUSE_Factory/GNOME:Factory.repo
zypper in libcanberra-gtk2-module

or if you don’t want to add the repo just install the needed package

zypper in https://download.opensuse.org/repositories/GNOME:/Factory/openSUSE_Factory/x86_64/libcanberra-gtk2-module-0.30-83.19.x86_64.rpm

Thanks, I had my hopes up but it’s not working, still much the same symptoms.

There ARE many instances of libcanberra-gtk-module.so under /usr/lib and /usr/lib64, plus I copied a few to parent directories like /usr but it still cannot find a single one of maybe half a dozen. V-0.30-14.3 was already installed, with your your suggestion I upgraded it to V-0.30-83.19. It ‘sems’ like the setup script is not doing a very good search or… this business is way over my head.

I wish someone would fix that LSB to include stick support


Gtk-WARNING **: Failed to load module “libcanberra-gtk-module.so”: libcanberra-gtk-module.so: cannot open shared object file: No such file or directory
loki_setup: Suspect size value for option option

loki_setup: Suspect size value for option option

Installing mimetypes…
Installing desktop menu entries…
Installing desktop icon…
Google Earth has caught signal 11.

We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:

/root/.googleearth/crashlogs/crashlog-59f763bd.txt

Please include this file if you submit a bug report to Google.

the file=

Major Version 5
Minor Version 2
Build Number 0001
Build Date Sep 1 2010
Build Time 11:25:42
OS Type 3
OS Major Version 4
OS Minor Version 13
OS Build Version 9
OS Patch Version 0
Crash Signal 11
Crash Time 1509385149
Up Time 0.396427

Stacktrace from glibc:
./libgoogleearth_free.so(+0xd090b)[0xf7f0b90b]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf7f69dd0]

=========


Want to try version 6.2.2?

From what I see it comes with bundled free horror stories!

Does it use LSB meaning NO stick support, or does it provide stick support by not using LSB?

https://productforums.google.com/forum/#!topic/earth/XNNJgyJBw3Y

why did you do that you shouldn’t mess with shared libraries in /usr you might have broken something
I’m guessing that version google earth is a 32 bit application so it would require the 32bit shared object try installing it

zypper in libcanberra-gtk2-module-32bit 

the above will work if you have the gnome:factory reposetory if not you need to add it

zypper ar -f https://download.opensuse.org/repositories/GNOME:/Factory/openSUSE_Factory/GNOME:Factory.repo

and don’t copy so files by hand there should be only 2 files a 64bit so in /usr/lib64 and a 32bit in /usr/lib

Files I copy are recorded and easily renamed/removed. They’re gone.

Yast shows the following installed:

libcanberra-gtk2-module 0.30-14.3-x86_64
libcanberra-gtk2-module-32bit 0.30-14.3-x86_64
libcanberra-gtk3-module 0.30-14.3-x86_64
libcanberra-gtk3-module-32bit 0.30-14.3-x86_64

.so files found

libcanberra-gtk-module.so /usr/lib64/gtk-2.0/modules 26.6kb
libcanberra-gtk-module.so /usr/lib64/gtk-3.0/modules 26.6kb
libcanberra-gtk-module.so /usr/lib /gtk-2.0/modules 21.7kb
libcanberra-gtk-module.so /usr/lib /gtk-3.0/modules 21.7kb
libcanberra-gtk3-module.so /usr/lib /gtk-3.0/modules 21.7kb
libcanberra-gtk3-module.so /usr/lib64/gtk-3.0/modules 26.6kb


zypper ar -f https://download.opensuse.org/repositories/GNOME:/Factory/openSUSE_Factory/GNOME:Factory.repo
Adding repository 'Development branch of GNOME (openSUSE_Factory)' .......................................................................................................................[done]
Repository 'Development branch of GNOME (openSUSE_Factory)' successfully added

URI         : http://download.opensuse.org/repositories/GNOME:/Factory/openSUSE_Factory/
Enabled     : Yes                                                                       
GPG Check   : Yes                                                                       
Autorefresh : Yes                                                                       
Priority    : 99 (default priority)                                                     

Repository priorities are without effect. All enabled repositories share the same priority.[/CODE


so then I installed:


zypper in libcanberra-gtk2-module-32bit
Retrieving repository ‘Development branch of GNOME (openSUSE_Factory)’ metadata -------------------------------------------------------------------------------------------------------------

New repository or package signing key received:

Repository: Development branch of GNOME (openSUSE_Factory)
Key Name: GNOME OBS Project GNOME@build.opensuse.org
Key Fingerprint: D3CAF513 5D0A8F97 AB539ED3 65A86F31 629FF0C2
Key Created: mar. 19 juil. 2016 05:39:47 EDT
Key Expires: jeu. 27 sept. 2018 05:39:47 EDT
Rpm Name: gpg-pubkey-629ff0c2-578df563

Do you want to reject the key, trust temporarily, or trust always? [r/t/a/? shows all options] (r): t
Retrieving repository ‘Development branch of GNOME (openSUSE_Factory)’ metadata …
Building repository ‘Development branch of GNOME (openSUSE_Factory)’ cache …
Loading repository data…
Reading installed packages…
‘libcanberra-gtk2-module-32bit’ is already installed.
There is an update candidate for ‘libcanberra-gtk2-module-32bit’, but it is from a different vendor. Use ‘zypper install libcanberra-gtk2-module-32bit-0.30-83.19.x86_64’ to iate.
Resolving package dependencies…

Nothing to do.



So then I executyed the suggested update:


zypper install libcanberra-gtk2-module-32bit-0.30-83.19.x86_64’ to iate.

zypper install libcanberra-gtk2-module-32bit-0.30-83.19.x86_64
Loading repository data…
Reading installed packages…
Resolving package dependencies…

The following package is going to be upgraded:
libcanberra-gtk2-module-32bit

The following package is going to change vendor:
libcanberra-gtk2-module-32bit openSUSE → obs://build.opensuse.org/GNOME

1 package to upgrade, 1 to change vendor.
Overall download size: 25.8 KiB. Already cached: 0 B. No additional space will be used or freed after the operation.
Continue? [y/n/…? shows all options] (y): y
Retrieving package libcanberra-gtk2-module-32bit-0.30-83.19.x86_64 (1/1), 25.8 KiB ( 21.7 KiB unpacked)
Retrieving: libcanberra-gtk2-module-32bit-0.30-83.19.x86_64.rpm …[done (1.0 KiB/s)]
Checking for file conflicts: …[done]
(1/1) Installing: libcanberra-gtk2-module-32bit-0.30-83.19.x86_64 …[done]
Additional rpm output:
warning: /var/cache/zypp/packages/GNOME_Factory/x86_64/libcanberra-gtk2-module-32bit-0.30-83.19.x86_64.rpm: Header V3 DSA/SHA1 Signature, key ID 629ff0c2: NOKEY
There are some running programs that might use files deleted by recent upgrade. You may wish to check and restart some of them. Run ‘zypper ps -s’ to list these programs.



At this point I rebooted and rescanned what was up:

Yast (installed packages) shows 1 change::
libcanberra-gtk2-module          0.30-14.3-x86_64
libcanberra-gtk2-module-32bit    0.30-83.19-x86_64
libcanberra-gtk3-module          0.30-14.3-x86_64
libcanberra-gtk3-module-32bit    0.30-14.3-x86_64

.so files found same as before, different order:

libcanberra-gtk-module.so     /usr/lib64/gtk-2.0/modules   26.6kb
libcanberra-gtk-module.so     /usr/lib  /gtk-2.0/modules   21.7kb
libcanberra-gtk-module.so     /usr/lib  /gtk-3.0/modules   21.7kb
libcanberra-gtk-module.so     /usr/lib64/gtk-3.0/modules   26.6kb
libcanberra-gtk3-module.so    /usr/lib  /gtk-3.0/modules   21.7kb
libcanberra-gtk3-module.so    /usr/lib64/gtk-3.0/modules   26.6kb

So one more relaunch of the google setup script:


setup.sh

Gtk-WARNING **: Failed to load module “libcanberra-gtk-module.so”: libcanberra-gtk-module.so: cannot open shared object file: No such file or directory
loki_setup: Suspect size value for option option
loki_setup: Suspect size value for option option
Installing mimetypes…
Installing desktop menu entries…
Installing desktop icon…



hung install but at this point the google installer dialog shows a START button

if I click it:


Google Earth has caught signal 11.
We apologize for the inconvenience, but Google Earth has crashed.
This is a bug in the program, and should never happen under normal
circumstances. A bug report and debugging data have been written
to this text file:
/root/.googleearth/crashlogs/crashlog-59f7d2c1.txt
Please include this file if you submit a bug report to Google.



the file:

Major Version 5
Minor Version 2
Build Number 0001
Build Date Sep  1 2010
Build Time 11:25:42
OS Type 3
OS Major Version 4
OS Minor Version 13
OS Build Version 9
OS Patch Version 0
Crash Signal 11
Crash Time 1509413569
Up Time 0.496405

Stacktrace from glibc:
./libgoogleearth_free.so(+0xd090b)[0xf7e8b90b]
linux-gate.so.1(__kernel_sigreturn+0x0)[0xf7ee9dd0]


So there, there seems to be a path problem, the setup file is over my head and I don't know how to upload it. If I paste it maes out the post so it's pasted as an answer to this with apologies :shame:

============================================================

#! /bin/sh
… see next



---------------------------------------------------------------------------------------------------------


#! /bin/sh
#
# Product setup script
#

if  -z "$DISPLAY" ]; then
    echo "You don't seem to be running an X server (no DISPLAY set)."
    echo "Google Earth and its installer both require X11."
    echo "Aborting..."
    exit 1
fi


# Go to the proper setup directory (if not already there)
cd `dirname $0`

# defaults
FATAL_ERROR="Fatal error, installer failed to run at all!"
# try to get root prior to running setup?
# 0: no
# 1: prompt, but run anyway if fails
# 2: require, abort if root fails
GET_ROOT=0
XSU_ICON=""
# You may want to set USE_XHOST to 1 if you want an X11 application to
# be launched with root privileges right after installation
USE_XHOST=0
# this is the message for su call, printf
SU_MESSAGE="You need to run this installation as the super user.
Please enter the root password."

NULL=/dev/null

# See if we have the XPG4 utilities (Solaris)
if test -d /usr/xpg4/bin; then
    PATH=/usr/xpg4/bin:$PATH
fi

# Return the appropriate architecture string
DetectARCH()
{
    status=1
    case `uname -m` in
        x86_64 | amd64)
        echo "amd64"
        status=0;;
        i?86 | i86*)
        echo "x86"
        status=0;;
        90*/*) 
        echo "hppa"
        status=0;;
        *)
        case `uname -s` in
            IRIX*)
            echo "mips"
            status=0;;
            AIX*)
            echo "ppc"
            status=0;;
            *)
            arch=`uname -p 2> /dev/null || uname -m`
            if test "$arch" = powerpc; then
                echo "ppc"
            else
                echo $arch
            fi
            status=0;;
        esac
    esac
    return $status
}

# Return the appropriate version string
DetectLIBC()
{
    status=1
      if  `uname -s` != Linux ]; then
          echo "glibc-2.1"
          return $status
      fi
      if  -f `echo /lib/libc.so.6* | tail -n 1` ]; then
          if fgrep GLIBC_2.1 /lib/libc.so.6* 2> $NULL >> $NULL; then
                  echo "glibc-2.1"
                  status=0
          else    
                  echo "glibc-2.0"
                  status=0
          fi        
      elif  -f /lib/libc.so.5 ]; then
          echo "libc5"
          status=0
      else
          echo "unknown"
      fi
      return $status
}

DetectOS()
{
    os=`uname -s`
    if test "$os" = "OpenUNIX"; then
        echo SCO_SV
    else
        echo $os
    fi
    return 0
}

# Detect the environment
arch=`DetectARCH`
libc=`DetectLIBC`
os=`DetectOS`

args=""

# Import preferences from a secondary script
if  -f setup.data/config.sh ]; then
    . setup.data/config.sh
elif  -f SETUP.DAT/CONFIG.SH\;1 ]; then
    # HP-UX and other systems unable to get LFN correctly
    . SETUP.DAT/CONFIG.SH\;1
fi

# Add some standard paths for compatibility
PATH=$PATH:/usr/ucb

# call setup with -auth when ran through su/xsu
auth=0
if  "$1" = "-auth" ]
then
  auth=1
  shift
fi

if  "$auth" -eq 1 ]
then
  # if root is absolutely required
  # this happens if xsu/su execs setup.sh but it still doesn't have root rights
  if  "$GET_ROOT" -eq 2 ]
  then
    # NOTE TTimo: this causes the following error message in some cases:
    # return: can only `return' from a function or sourced script
    # BUT: in other cases, the return is legit, if you replace by an exit call, it's broken
    exit 1
  fi
fi

# Find the installation program
# try_run -absolute] -fatal] INSTALLER_NAME [PARAMETERS_PASSED]
#   -absolute option: if what you are trying to execute has an absolute path
#   NOTE: maybe try_run_absolute would be easier
#   -fatal option: if you want verbose messages in case
#      - the script could not be found
#      - it's execution would fail
#   INSTALLER_NAME: setup.gtk or setup, etc
#   PARAMETERS_PASSED: additional arguments passed to the setup script
try_run()
{
    absolute=0
    if  "$1" = "-absolute" ]; then
      absolute=1
      shift
    fi

    fatal=0
    # older bash < 2.* don't like == operator, using =
    if  "$1" = "-fatal" ]; then
      # got fatal
      fatal=1
      shift
    fi

    setup=$1
    shift

    # echo "$os" "$arch"
    
    # First find the binary we want to run
    failed=0
    if  "$absolute" -eq 0 ]
    then
      setup_bin="setup.data/bin/$os/$arch/$libc/$setup"
      # trying $setup_bin
      if  ! -f "$setup_bin" ]; then
          setup_bin="setup.data/bin/$os/$arch/$setup"
            # libc dependant version failed, trying again
          if  ! -f "$setup_bin" ]; then
              failed=1
          fi
      fi
      if  "$failed" -eq 1 ]; then
          if  "$fatal" -eq 1 ]; then
              echo "This installation doesn't support $libc on $os / $arch"
              echo "(tried to run $setup)"
              echo "$FATAL_ERROR"
          fi
          return $failed
      fi

      # Try to run the binary ($setup_bin)
      # The executable is here but we can't execute it from CD
      # NOTE TTimo: this is dangerous, we also use $setup to store the name of the try_run

      #setup="$HOME/.setup$$"
      #rm -f "$setup"
      #cp "$setup_bin" "$setup"
      #chmod 700 "$setup"
      #trap "rm -f $setup" 1 2 3 15

      # Google Earth isn't on a CD (and if it was, makeself would have
      # unpacked it to /tmp anyhow), so don't copy. --ryan.
      setup="$setup_bin"
    fi
    # echo Running "$setup" "$@"
    if  "$fatal" -eq 0 ]; then
        "$setup" "$@"
        failed="$?"
    else
        "$setup" "$@" 2>> $NULL
        failed="$?"
    fi
    if  "$absolute" -eq 0 ]
    then
      # don't attempt removal when we are passed an absolute path
      # no, I don't want to imagine a faulty try_run as root on /bin/su

      # Google Earth installer didn't copy here, so don't unlink.
      ignorethisline=1 #rm -f "$setup"
    fi
    return "$failed"
}

# if we have not been through the auth yet, and if we need to get root, then prompt
if  "$auth" -eq 0 ] &&  "$GET_ROOT" -ne 0 ]
then
  GOT_ROOT=`id -u`
  if  "$GOT_ROOT" != "0" ]
  then
    if  "$USE_XHOST" -eq 1 ]; then
        xhost +127.0.0.1 2> $NULL > $NULL
    fi
    try_run xsu -e -a -u root -c "sh `pwd`/setup.sh -auth" $XSU_ICON
    status="$?"
    # echo "got $status"
    # if try_run successfully executed xsu, it will return xsu's exit code
    # xsu returns 2 if ran and cancelled (i.e. the user 'doesn't want' to auth)
    # it will return 0 if the command was executed correctly
    # summing up, if we get 1, something failed
    if  "$status" -eq 0 ]
    then
      # the auth command was properly executed
      exit 0
    elif  "$status" -eq 1 ]
    then
      # xsu wasn't found, or failed to run
      # if xsu actually ran and the auth was cancelled, $status is 2
      # try with su
      printf "$SU_MESSAGE
"
      try_run -absolute /bin/su root -c "export DISPLAY=$DISPLAY;sh `pwd`/setup.sh -auth"
      status="$?"
      if  "$status" -eq 0 ]; then
        # the auth command was properly executed
        exit 0
      else
        exit 1
      fi
    elif  "$status" -eq 3 ]
    then
      # the auth failed or was canceled
      # we don't want to even start the setup if not root
      echo "Please run this installation as the super user"
      exit 1
    fi
    # continue running as is
  fi
fi

# Try to run the setup program. GTK+v2 first, if that fails, try GTK+v1, then try a stdio thing.
if  ! -z "$SETUP_PROGRAM" ]; then
    try_run "$SETUP_PROGRAM" $args $*
    exit $?
fi

try_run setup.gtk2 $args $*
status=$?
if  $status -eq 2 ] ||  $status -eq 127 ]; then  # setup.gtk couldn't connect to X11 server - ignore
    try_run setup.gtk $args $*
    status=$?
    if  $status -eq 2 ] ||  $status -eq 127 ]; then  # setup.gtk couldn't connect to X11 server - ignore
        try_run setup $args $* || {
            if  $status -ne 2 ]; then
                echo "The setup program seems to have failed on $arch"
                echo
                echo $FATAL_ERROR
            fi
            status=1
        }
    fi
fi
exit $status





This may be a NOT OpenSuse issue after all. Once I had seen that an install to a somewhat tweaked Debian-8 went alright I moved up to Debian-9 because after all what makers all this critical from MY perspective is the potential INability to continue using Ge-Flt-Sim with Linux AND stick support in years to come. Having cleared the subject Debian-8 install I tried the install to Debian-9 and got weird results there too. The suspected setup script is unable to find some other fles in this case, again it’s files that are staring it right in the face. There the ‘missing’ files setup.gtk2 and setup.gtk are actually part of the package that the installer itself unpacks!


# fixit/setup.sh
/0/sa14/comp-gen/xtra-progs/fixit/setup.sh: 216: /0/sa14/comp-gen/xtra-progs/fixit/setup.sh: setup.data/bin/Linux/amd64/setup.gtk2: not found
/0/sa14/comp-gen/xtra-progs/fixit/setup.sh: 216: /0/sa14/comp-gen/xtra-progs/fixit/setup.sh: setup.data/bin/Linux/amd64/setup.gtk: not found
The setup program seems to have failed on amd64
Fatal error, installer failed to run at all!

So, methinks maybe I should try this in a general Linux forum on usenet. The ideal solution would of course be to get stick support into LSB, is anyone maintainingthat package at all? I mean it’s been many years now…

Maybe I’m missing something, but I downloaded the rpm from https://www.google.nl/earth/download/gep/agree.html to ~/rpms in my homedir
Next I opened a terminal window, cd’d into that folder and did

su -c 'zypper in ./google-earth-pro-stable-current.x86_64.rpm'

The rpm installed properly, no warnings or errors. I started GE and it works like it should.

My issue with versions > 5.2 is not installation but that the Flt-Sim has no stick support, which is capital nonsense. As far as I know the reason for this is that said later versions use the LSB package and that it has no stick support for some obscure reason and that this shortcoming has never been addressed since. Anyway, I did want to get the ‘latest and greatest’ aboard just to be sure…


su -c 'zypper in ./google-earth-pro-stable-current.x86_64.rpm'
Password: 
Specified local path does not exist or is not accessible.
Problem retrieving the specified RPM file:
Malformed URI: 
Please check whether the file is accessible.
Problem with the RPM file specified as './google-earth-pro-stable-current.x86_64.rpm', skipping.
No valid arguments specified.

I found no Pro version in the online repos that I have but did manage to get the

google-earth-pro-stable-current.x86_64.rpm
from
https://www.google.com/earth/download/gep/agree.html

Enabled in Yast the local folder hosting it after the download and installed it no problem.

Upon launching all I get is a graphical disaster (too bad I cannoty upload a screenshot) but I can get to the menu line to select “Enter Flight Simulator” or such. The resulting Flt-Sim setup dialog has a button “Joystick Enabled” which is ghosted in ALL versions since 5.2 as far as I know. The stick IS there


lsusb
Bus 006 Device 002: ID 044f:b10a ThrustMaster, Inc. T.16000M Joystick

and it works fine with Ge-5.2 on OpenSuse-13.2.

None says an old version of software will run on a newer OS.

Seems like your problem is with Google and their depreciation of the flight stick feature

For what it is worth I still have the old 5.2 here and it runs fine on 42.2. But then TW is cutting edge and far a head of that version

My problem is with the idiots that wrote LSB without stick support. Google just took what was offered them by way of standardisation in Linux. Mind you after all these YEARS as far as I’m concerned BOTH groups should have cleared up the problem …maybe cooperatively.

Did you install the 5.2 rpm that’s floating around or the compressed google package? I haven’t tried Leap but if it works there then I see no reason why it shouldn’t on TW. I might give it a shot if all else fails, that would give a couple more years of use.

No I have had the same GE installed over several OS version since it lives in home and I keep home between OSs and is not a system install no need to reinstall it.

I’m gettin’ there I think. Did a Leap freshie. Not wanting to fall short of anything I cut a 200gb partition and poupulated it with 60gb worth of 42.3. The install attempt still bombed just like on TW but when I tried to just launch a binary on a 13.2 mounted for the occasion it took right off! It didn’t on TW but even this much buys me a couple of years. Next I’ll try to repeat the same thing on a fresh TW although success there would be no guarantee of longer term survival.

Did that, no change, Ge-5.2 will neither install on TW nor run if just invoked as a transplanted binary. However I also noticed that a Leap install with just the default packages isn’t enough either. I padded out the Leap install with every available package from the DVD minus half a dozen dependency failures. The same didn’t do it on TW however.

I took it one step further without success.

Imported into Yast the successful Leap-423 package list and installed from the TW repos with the fewest possible dependency related changes (in all I think maybe a dozen were not installed).

I got the same error message if I tried to run an existing Ge-5.2 binary OR if I tried to install the Google package, and I got it either as user OR as root:

Invalid MIT-MAGIC-COOKIE-1 key

Haven’t got the faintest clue as to what is means, or how to work around the problem any further.