CosmosEx 3.0 issues - can't get it to work [SOLVED]

[dhedberg]

Hi,

I received my CosmosEx 3.0 a while ago and today I finally had the time to have a go with it. As I'm going to use with a TT030 I figured the SCSI interface would be preferable. So time to reconfigure the CosmosEx to use the SCSI interface. I powered it down, changed the DIP switch on the board, and plugged in just the network and power cables.

When powering it on, it got stuck and never booted (waited 5 minutes). I knew that this could happen as the problem seems to be fairly common and described in more detail along with a solution here: https://www.youtube.com/watch?v=JMjCRRTZ3NM

First I verified that the CosmosEx was still operational by switching back to ACSI mode, and it booted just fine.
Back to SCSI mode, and it got stuck again. So, I did what was described in the youtube video.
I connected to the CosmosEx via ssh, but then I discovered that it had no /ce/ce_force_flash.sh file which you're supposed to run.
I figured I had to update the CosmosEx first, so I ran /ce/ce_update.sh.

Code: Select all

root@CosmosEx3:/ce# ./ce_update.sh
Terminating cesuper.sh
Terminating cosmosex not needed, no instance is running

Updating CosmosEx from internet, this will take a while.
DO NOT POWER OFF THE DEVICE!!!

Will try to install missing git
Get:1 http://archive.raspberrypi.org/debian stretch InRelease [25.4 kB]
Get:2 http://raspbian.raspberrypi.org/raspbian stretch InRelease [15.0 kB]
Get:3 http://archive.raspberrypi.org/debian stretch/main armhf Packages [192 kB]
Get:4 http://archive.raspberrypi.org/debian stretch/ui armhf Packages [45.0 kB]
Get:5 http://raspbian.raspberrypi.org/raspbian stretch/main armhf Packages [11.7 MB]
Get:6 http://raspbian.raspberrypi.org/raspbian stretch/contrib armhf Packages [56.9 kB]
Get:7 http://raspbian.raspberrypi.org/raspbian stretch/non-free armhf Packages [98.9 kB]
Fetched 12.1 MB in 18s (641 kB/s)
Reading package lists... Done
Reading package lists... Done
Building dependency tree
Reading state information... Done
The following additional packages will be installed:
  git-man liberror-perl
Suggested packages:
  git-daemon-run | git-daemon-sysvinit git-doc git-el git-email git-gui gitk gitweb git-arch git-cvs git-mediawiki
  git-svn
The following NEW packages will be installed:
  git git-man liberror-perl
0 upgraded, 3 newly installed, 0 to remove and 140 not upgraded.
Need to get 4,857 kB of archives.
After this operation, 26.4 MB of additional disk space will be used.
Get:1 http://raspbian.playstar.se/raspbian stretch/main armhf liberror-perl all 0.17024-1 [26.9 kB]
Get:2 http://raspbian.playstar.se/raspbian stretch/main armhf git-man all 1:2.11.0-3+deb9u6 [1,436 kB]
Get:3 http://raspbian.playstar.se/raspbian stretch/main armhf git armhf 1:2.11.0-3+deb9u6 [3,393 kB]
Fetched 4,857 kB in 1s (2,908 kB/s)
Selecting previously unselected package liberror-perl.
(Reading database ... 34710 files and directories currently installed.)
Preparing to unpack .../liberror-perl_0.17024-1_all.deb ...
Unpacking liberror-perl (0.17024-1) ...
Selecting previously unselected package git-man.
Preparing to unpack .../git-man_1%3a2.11.0-3+deb9u6_all.deb ...
Unpacking git-man (1:2.11.0-3+deb9u6) ...
Selecting previously unselected package git.
Preparing to unpack .../git_1%3a2.11.0-3+deb9u6_armhf.deb ...
Unpacking git (1:2.11.0-3+deb9u6) ...
Setting up git-man (1:2.11.0-3+deb9u6) ...
Setting up liberror-perl (0.17024-1) ...
Processing triggers for man-db (2.7.6.1-2) ...
Setting up git (1:2.11.0-3+deb9u6) ...
W: --force-yes is deprecated, use one of the options starting with --allow instead.
git version 2.11.0
doing git pull...
fatal: Not a git repository (or any of the parent directories): .git
fatal: Not a git repository (or any of the parent directories): .git
doing git fetch...
Initialized empty Git repository in /ce/.git/
remote: Enumerating objects: 137, done.
remote: Counting objects: 100% (137/137), done.
remote: Compressing objects: 100% (120/120), done.
remote: Total 137 (delta 23), reused 84 (delta 7), pack-reused 0
Receiving objects: 100% (137/137), 1.71 MiB | 2.76 MiB/s, done.
Resolving deltas: 100% (23/23), done.
From https://github.com/atarijookie/ce-atari-releases
 * [new branch]      jessie     -> origin/jessie
 * [new branch]      master     -> origin/master
 * [new branch]      stretch    -> origin/stretch
 * [new branch]      yocto      -> origin/yocto
Branch stretch set up to track remote branch stretch from origin.
Switched to a new branch 'stretch'
From https://github.com/atarijookie/ce-atari-releases
 * branch            stretch    -> FETCH_HEAD
Already up-to-date.
XILINX firmware writing script.
Detected XC9572 chip, now will detect if it's ACSI or SCSI
Detected XC9572 chip and SCSI interface, will write firmware
XSVF Player v5.01, Xilinx, Inc.
XSVF file = /ce/update/xlnx2s.xsvf
ERROR:  TDO mismatch and exceeded max retries
ERROR at or near XSVF command #382.  See line #382 in the XSVF ASCII file.
Execution Time = 157.7 seconds
----------------------------------

>>> Updating Hans - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash HANS <<<

Using Parser : Intel HEX
Serial Config: 57600 8E1
read_byte: Success
flash_stm32: stm32.c:90: stm32_read_byte: Assertion `0' failed.
Aborted
----------------------------------

>>> Updating Franz - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash FRANZ <<<

Using Parser : Intel HEX
Serial Config: 57600 8E1
read_byte: Success
flash_stm32: stm32.c:90: stm32_read_byte: Assertion `0' failed.
Aborted

>>> Updating Franz - END
----------------------------------
core_freq is ok in /boot/config.txt

Update done, you may start the /ce/ce_start.sh now!

root@CosmosEx3:/ce#  

Executing /ce/ce_update.sh tried to flash the chips but failed. It was good to see that the choice of SCSI interface was properly detected though.
Anyways, the update gave me the missing /ce/ce_force_flash.sh file. Great!
So went ahead and executed ce_force_flash.sh with the following results:

Code: Select all

root@CosmosEx3:/ce# ./ce_force_flash.sh
Forcing writing of flash of all chips!
ce_update.sh started at : Fri 14 Aug 18:16:38 UTC 2020
Terminating cesuper.sh
Terminating cosmosex not needed, no instance is running

Updating CosmosEx from internet, this will take a while.
DO NOT POWER OFF THE DEVICE!!!

git version 2.11.0
git version 2.11.0
doing git pull...
HEAD is now at ec83f84 CR/LF EOL for readme.txt
From https://github.com/atarijookie/ce-atari-releases
 * branch            stretch    -> FETCH_HEAD
Already up-to-date.
XILINX firmware writing script.
Detected XC9572 chip, now will detect if it's ACSI or SCSI
Detected XC9572 chip and ACSI interface, will write firmware
XSVF Player v5.01, Xilinx, Inc.
XSVF file = /ce/update/xlnx2a.xsvf
SUCCESS - Completed XSVF execution.
Execution Time = 46.4 seconds
----------------------------------

>>> Updating Hans - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash HANS <<<

Using Parser : Intel HEX
Serial Config: 57600 8E1
read_byte: Success
flash_stm32: stm32.c:90: stm32_read_byte: Assertion `0' failed.
Aborted
----------------------------------

>>> Updating Franz - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash FRANZ <<<

Using Parser : Intel HEX
Serial Config: 57600 8E1
read_byte: Success
flash_stm32: stm32.c:90: stm32_read_byte: Assertion `0' failed.
Aborted

>>> Updating Franz - END
----------------------------------
check_and_flash_chips.sh finished
core_freq is ok in /boot/config.txt

Update done, you may start the /ce/ce_start.sh now!

ce_update.sh finished at: Fri 14 Aug 18:17:38 UTC 2020
root@CosmosEx3:/ce#  

As can be seen the XILINX firmware was flashed, but for some reason ACSI mode was detected even though I had the switch in SCSI mode.
For some reason the flashing of HANS and FRANZ failed.

Anyways, power cycled the unit to see if it would boot... but it did not.
So I once again connected to the unit via ssh and this is what I found:

Code: Select all

root@192.168.1.101's password:
Linux CosmosEx3 4.14.34-v7+ #1110 SMP Mon Apr 16 15:18:51 BST 2018 armv7l
CosmosEx running on Raspbian Stretch.
To enable USB keyboard in linux console, press:
Left Ctrl + Left Alt + Right Ctrl + Right Alt
root@CosmosEx3:~# cd ..
root@CosmosEx3:/# cd ce
root@CosmosEx3:/ce# ls -l
total 64
drwxr-xr-x 5 root root  4096 Aug 14 18:25 app
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ceboot.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_conf.sh
-rwxr-xr-x 1 root root  1104 Oct 16  2016 ce_firstfw.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_force_flash.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_running.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_start.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_stop.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 cesuper.sh
-rwxr-xr-x 1 root root     0 Aug 14 18:17 ce_update.sh
-rw-r--r-- 1 root root   263 Aug 14 18:10 cosmosex.service
drwxr-xr-x 2 root root  4096 May 10  2018 firstfw
-rw-r--r-- 1 root root   809 Aug 14 18:10 initd_cosmosex
-rwxr-xr-x 1 root root     0 Aug 14 18:17 install_ffmpeg.sh
-rw-r--r-- 1 root root 35150 Aug 14 18:10 LICENSE
drwxr-xr-x 2 root root  4096 Aug 14 18:25 settings
-rwxr-xr-x 1 root root     0 Aug 14 18:17 time.sh
drwxr-xr-x 2 root root  4096 Aug 14 18:17 update
-rwxr-xr-x 1 root root     0 Aug 14 18:17 whichdistro.sh
root@CosmosEx3:/ce#

All the .sh files are 0 bytes, including ce_force_flash.sh that I just executed! What the heck? How did that happen?
Now, I have no clue what to do to get this thing to work. Any suggestions would be most welcome. Until then, I'll put it on a shelf.

Thanks,
Daniel

[ube]

We fixed it. This is the oneliner to restore everything..

Code: Select all

cd / && mv /ce /_ce-orig && git clone https://github.com/atarijookie/ce-atari-releases.git /ce && cd /ce && git checkout stretch && reboot

[dhedberg]

We fixed it. This is the oneliner to restore everything..

So thankful that you took your time and helped me out on this matter Ube! You saved my CosmosEx! Thanks!

[ube]

The professional in me wants to do a root cause analysis. But I’ll leave it to this.

[whiskey]

The professional in me wants to do a root cause analysis. But I’ll leave it to this.

any major incident should trigger a related problem ticket!

[GoldenGrahams]

Hi guys,

I'm trying to change my CosmosEx from ACSI mode to SCSI mode so I an use it in my Falcon, however the documented method of removing the dip switch and adding the two resistor networks is having no affect. Unfortunately my /ce folder doesn't have the ce_force_flash.sh file so I cannot force the switch from ACSI to SCSI in that way. I have therefore tried executing ce_update.sh to get a later release which might include ce_force_flash.sh . However when I execute ce_update.sh I get a number of errors as can be seen in the output pasted below. It seems my CEx is missing 'git' and is therefore unable to download the necessary files to update.

Note that the 'one liner' in this thread does not work for me because 'git' is reported as missing.

Please could someone advise on...

a) How get 'git' and resolve the errors reported below
b) How to completely restore my CEx to factory settings
c) How to switch from ACSI mode to SCSI mode when the normal 'remove DIP switch' method doesn't work.

Note that the version config of my CEx is currently reported as follows which strangely suggests that everything is up to date...

Main App 2019-07-22
Franz 2016-01-17
Hans 2016-01-08
Xilinx 2015-12-06

Any help/guidance on this whatsoever is very much appreciated.

Code: Select all

root@ce:/ce# ./ce_update.sh
Terminating cesuper.sh
Terminating cosmosex - SIGINT
Waiting for cosmosex to terminate nicely...
cosmosex terminated

Updating CosmosEx from internet, this will take a while.
DO NOT POWER OFF THE DEVICE!!!

Will try to install missing git
./ce_update.sh: line 44: apt-get: command not found
./ce_update.sh: line 45: apt-get: command not found
git is still missing, will use wget
wget: not an http or ftp url: https://github.com/atarijookie/ce-atari-releases/a                                                                                                                                                             rchive/yocto.zip
unzip: can't open /tmp/yocto.zip[.zip]
XILINX firmware writing script.
Detected XC9572 chip, now will detect if it's ACSI or SCSI
Detected XC9572 chip and ACSI interface, will write firmware
XSVF Player v5.01, Xilinx, Inc.
XSVF file = /ce/update/xlnx2a.xsvf
ERROR:  Cannot open file /ce/update/xlnx2a.xsvf
cp: can't stat '/ce/update/xlnx2a.version': No such file or directory
----------------------------------

>>> Updating Hans - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash HANS <<<

Intel HEX ERROR: System Error
/ce/update/hans.hex: No such file or directory

cp: can't stat '/ce/update/hans.version': No such file or directory
----------------------------------

>>> Updating Franz - START
Will use serial port:  /dev/ttyAMA0

stm32flash - http://stm32flash.googlecode.com/
           - with the fix for DS1818

>>> Will flash FRANZ <<<

Intel HEX ERROR: System Error
/ce/update/franz.hex: No such file or directory


>>> Updating Franz - END
----------------------------------
cp: can't stat '/ce/update/franz.version': No such file or directory
core_freq is ok in /boot/config.txt
ln: /lib/systemd/system/ctrl-alt-del.target: No such file or directory
ln: /etc/systemd/system/getty.target.wants/getty@tty1.service: No such file or d                                                                                                                                                             irectory

Update done, you may start the /ce/ce_start.sh now!

[sporniket]

Hum, maybe a "sudo apt-get install git" would solve your problem ? On a previously pasted logs, it says it is running on Raspbian.

[GoldenGrahams]

Many thanks for your suggestion, however I just get 'sudo: command not found' when I try executing the command you suggested from the root@ce:/ce# command prompt . Do you have any further advice as to how/where I should be executing that command?

[dhedberg]

What CosmosEx do you have and what Linux distro is installed on it?

[sporniket]

ho, I did not noticed, "root@ce:/ce#", you are already root... just type "apt-get install git"

[GoldenGrahams]

ho, I did not noticed, "root@ce:/ce#", you are already root... just type "apt-get install git"

Thanks. Similar problem I get this...

apt-get: command not found

[GoldenGrahams]

What CosmosEx do you have and what Linux distro is installed on it?

Hi, my CosmosEx is version 2.0 (the SCSI version) bought direct from Jookie in November 2015. I've no idea which Linux distro is installed on it, how do I check?

[sporniket]

apt-get: command not found

Well, I've run out of simple procedure to get going. At this point I wonder whether a lot of the usual programs have been erased by an update process gone wrong, and whether one should just get the micro sd cards and restore it from a working image...

[sporniket]

Reading the log again, I just noticed that the script try to download an archive, the log it is written
"https://github.com/atarijookie/ce-atari-releases/a[... a lot of spaces...]rchive/yocto.zip", instead of "https://github.com/atarijookie/ce-atari ... /yocto.zip", and thus wget rejected the target url as invalid

Check the "ce_update.sh" file for a call to the "wget" program to download said archive "yocto.zip". Anyway, I could get it on my browser and could see that it contains the firmware update.

[GoldenGrahams]

apt-get: command not found

Well, I've run out of simple procedure to get going. At this point I wonder whether a lot of the usual programs have been erased by an update process gone wrong, and whether one should just get the micro sd cards and restore it from a working image...

Thanks, I'm glad you suggested re imaging the micro sd card of the Raspberry Pi within my CEx. I'm beginning to think that's the best/easiest option for me to restore.

Please can you confirm what the latest version of this image I can use? Do I have to restore exactly the same version of SD image as what my CEx shipped with? Or can I use a later version. My CEx is version 2.0 purchased November 2015. I understand there is now a version 3.0 of CEx now available, am I able to use an SD image for CEx version 3.0 in my CEx version 2.0...? Obviously I would like to use the best and latest version available for my CEx 2.0. The only place I know where such images are available is at the following address, but not which one I should use...

http://joo.kie.sk/?page_id=474

Many thanks.

[sporniket]

Please check my other message about the weird url for the 'yocto.zip' archive, before resorting to reimaging the sd cards.

About which image to use, I cannot answer for sure as I don't own one. That said, given there are mention of the "yocto" and reading the page you mention, I infer that your CEx is based on Raspberry 1, thus I would stick to the "yocto" card image.

[mikro]

Yup, just take the latest SD card image and write it to the SD card. Yocto is super old, deprecated, unsupported etc.

I can confirm that writing http://joo.kie.sk/cosmosex/download/ce_ ... -08-13.zip for a v2 device is totally OK.

[GoldenGrahams]

Thanks @sporniket, sorry that large space didn't actually exist in the output from my ce_update.sh. I must've accidentally introduced those spaces as I pasted it into this thread. I think the wget command in my ce_update.sh file is failing because the wget command does not support https. The location of the yocto.zip file appears to have changed in recent years from http to https which causes the error listed above.

So instead I downloaded to yocto.zip file using my PC and put it on a USB stick and performed the update using the usb update method which eventually worked fine for me and updated all 4 components of my CosmosEx to the very latest versions. What's more, as part of the update it also finally recognised that the jumper was set to SCSI mode and so switched from ACSI to SCSI mode properly too. Well happy

After a fair bit of testing the CosmosEx now works with my Falcon030. It also appears to work okay in 060 mode with my CT60e enabled too.

Thank you also to @mikro. I will now try that later Linux distro that you linked me to. What extra features in the CosmosEx are made available by using that later Linux distro?

Also can anyone please tell me what the differences are (in terms of features and functionality) between version 2.0 of the CosmosEx that Jookie produced a few years ago (which I have) and the CosmosEx version 3.0 that's available to buy elsewhere now?

Thanks.

[mikro]

What extra features in the CosmosEx are made available by using that later Linux distro?

Not much of visible (user) features but if something goes wrong, people will be more likely able to help you. :)

As for v2 vs. v3 features: it boots faster, offers true LCD display with customised text, it better fits into Falcons, it has built-in wifi and bluetooth, IKBD sandwich works immediately after power up, ...

[GoldenGrahams]

Many thanks @mikro , the raspbian_stretch distro you linked me to does indeed work fine on my v2.0 CosmosEx. I've had no problems with fitting the v2.0 CEx in my Falcon so far but I can see that the v3.0 CEx is more compact and would not take up quite as much space. The other extra features of v3.0 do not interest me too much.

The big problem I having with my CEx right now is the 'Shared Drive' is not working when mapped to my NAS drive (which is a 'D-Link ShareCentre DNS-320L'). This worked perfectly before when I used the same CEx (with Yocto distro) in my STE with the same NAS drive. I have no problems whatsoever when mapping the 'Shared Drive' to a share on my Windows10 PC (on the same network). It just seems to have a problem with mapping to my NAS.

From my Windows10 PC I am able to map a 'shared network drive' to my NAS at //192.168.1.4/Atari however I am not able to map a 'Shared Drive' to the same address form my CosmosEx/Falcon.

The error given (in the MOUNT.ERR file in the P: drive on my Falcon) is 'Host is down'. The full error file reads as follows...

mount() failed : Host is down
source : '//192.168.1.4/Atari'
type : 'cifs'
target : 'mnt/shared'
options : 'gid=1000,forcegid,uid=1000,forceuid,username=Atari,password=Atari,_ne'

I can't understand how this worked perfectly under my previous setup in my STE (with yocto instead of raspbian) but not in my Falcon. I have made sure my share paths, share name, username and password are as simple as possible. I've made sure that the 'cifs' protocol is enabled on my NAS. The file system on my NAS is Linux based I believe.

Here's the 'Shared Drive' config on the CEx side of things...

Enabled: [*]
Sharing protocol [*] Samba / cifs / windows
IP address of server [192.168.1.4 ]
Shared folder path on server [Atari ]
Username [Atari ]
Password [***** ]

Has anybody else ever experienced or heard of issues like this... and can offer any suggestions?

Many thanks.

[mjvans]

network settings on CEx?
Fill in DNS and Gateway to your router.
Yes all NAS boxes are some Linux distribution. I had trouble connecting to a Windows server. Every time password had to be typed in on CEx config and saved again. I changed to a linux Open Media Vault server and it works stable.
Shared drive settings look ok. I also also use Samba/cifs/windows option.

[GoldenGrahams]

network settings on CEx?
Fill in DNS and Gateway to your router.
Yes all NAS boxes are some Linux distribution. I had trouble connecting to a Windows server. Every time password had to be typed in on CEx config and saved again. I changed to a linux Open Media Vault server and it works stable.
Shared drive settings look ok. I also also use Samba/cifs/windows option.

The Network Settings on my CEx are the default settings which are as follows...

Hostname [cosmosex]
DNS [ ]
Ethernet
Use DHCP [*]
IP Address [192.168.1.105]
Mask [255.255.255.0]
Gateway [ ]
Wifi
Enable [*]
WPA SSID [ ]
WPA PSK [ ]
Use DHCP [*]
IP address [0.0.0.0 ]
Mask [0.0.0.0]
Gateway [ ]

My network and router are standard setup so I'm pretty sure that the DNS and Gateway to my router are both 192.168.1.1 . I have tried submitting these values in the Network Settings screen as you suggested however the new values I entered for DNS and Gateway have returned to blank and do not appear to have saved. I think this is because I have 'Use DHCP' set to true.

My CosmosEx definitely has a network connection and an internet connection because it was successfully able to update its own software using the 'online' method. I have also tried logging into the ssh console (as the root user) and pinging the IP address of my NAS (192.168.1.4) the ping is successful and confirms that my CosmosEx can 'see' my NAS.

I've still no idea why this Shared Drive setup is returning 'Host is down'. Any other thoughts guys? or things to try to narrow the problem down... or logs to check etc...?

Thank you.

[mjvans]

Yes you can ping withouth gateway or DNS because your ip range (adress and mask) is good. But your CEx can't find your server by dns.
Something goes wrong with DHCP server it doesn't provide a gateway or dns.
Try running cmd ipconfig /all on your W10 pc.
You should see DNS and Gateway . Take first DNS.
Assuming your gateway is 192.168.1.1
Assuming DHCP range is 192.168.100..255
You should check DHCP range on DHCP server probably in the router.
Uncheck Use DHCP on CEx
Use an ip adres not in DHCP range because another dhcp client could get the 105 you got earlier and double ip's are forbidden on a network.
something like 192.168.1.99
mask 255.255.255.0
gateway 192.168.1.1
dns 192.168.1.1 or dns found on W10 pc.

Another way is if you do use get ip adress automaticaly from DHCP your DHCP has to provide DNS and Gateway adresses. You should fix that in your DCHP server.

[GoldenGrahams]

Yes you can ping withouth gateway or DNS because your ip range (adress and mask) is good. But your CEx can't find your server by dns.
Something goes wrong with DHCP server it doesn't provide a gateway or dns.
Try running cmd ipconfig /all on your W10 pc.
You should see DNS and Gateway . Take first DNS.
Assuming your gateway is 192.168.1.1
Assuming DHCP range is 192.168.100..255
You should check DHCP range on DHCP server probably in the router.
Uncheck Use DHCP on CEx
Use an ip adres not in DHCP range because another dhcp client could get the 105 you got earlier and double ip's are forbidden on a network.
something like 192.168.1.99
mask 255.255.255.0
gateway 192.168.1.1
dns 192.168.1.1 or dns found on W10 pc.

Another way is if you do use get ip adress automaticaly from DHCP your DHCP has to provide DNS and Gateway adresses. You should fix that in your DCHP server.

Many thanks @mjvans for your guidance on this. Running ipconfig /all on my Windows10 PC shows the following about my home network...

DNS = 192.168.1.1
DHCP = 192.168.1.1
Default Gateway = 192.168.1.1
Subnet Mask = 255.255.255.0

I checked the DHCP setup on my home WiFi Hub. The DHCP range was setup as 192.168.1.10-254 . I have now changed this to 192.168.1.100-254 as you've advised.

I also updated the Network Settings on my CosmosEx as advised as below...
Note: strangely I have to enable the 'Use DHCP' option within the WiFi section else the form will not save and I get the message "Some wifi network address has invalid format or is empty, Please fix this and try again".

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
Hostname [CosmosEx]
DNS [192.168.1.1]

Ethernet
Use DHCP []
IP address [192.168.1.99]
Mask [255.255.255.0]
Gateway [192.168.1.1]

Wifi
Enable []
WPA SSID [ ]
WPA PSK [ ]

Use DHCP [*]
IP address [0.0.0.0]
Mask [0.0.0.0]
Gateway [ ]
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

I've saved the above values and then restarted my Falcon and CEx but unfortunately the message in the 'MOUNT.ERR' file in the shared 'P' drive on my Falcon now reads 'mount() failed : Network is unreachable'. Also now when I try to ping the ip address I just set for my CEx (192.168.1.99) from my Windows10 PC I get 'Destination host unreachable'. So for some reason it seems like something is not happy with that static '99' ip address that I set for my CosmosEx.

Meanwhile, regarding the other approach you suggested with enabling DHCP on the CosmosEx. The DHCP section on my home WiFi Hub appears to have no provision for 'fixing' the DNS and Gateway addresses.

Everything else on my CosmosEx seems to work okay, it's just this 'Shared Drive' feature (the main feature in my view!) that I'm struggling with.
Your help is greatly appreciated. Do you have any other suggestions for me or things to test/try? do I need to reboot my home WiFi Hub perhaps?

Many thanks.

[mjvans]

Euh you talk about wifi but wifi looks disabled on CEx.
The settings above work only whith network cable attached.
You could try with a network cable first. When this works try fixing your wifi settings.