Strange network/WiFi issue

[bpa]

I logged into the Radio using Putty successfully but thought it would be easier to find the log file with WinSCP. When I try to connect I get an error message

Error getting name of current remote directory
Command 'pwd'
failed with return code 0 and error message
-sh: groups: not found.

"pwd" works OK for me after ssh'ing into a Touch.

Latest log file is in /var/log/messages - there is also /var/log/messages.0 which I think is created from messages about every 10 mins.

[slartibartfast]

"pwd" works OK for me after ssh'ing into a Touch.

Latest log file is in /var/log/messages - there is also /var/log/messages.0 which I think is created from messages about every 10 mins.

The error seems to be a red herring. If I close the error message I can see the file system. Enabling debug for net.http is extremely verbose. I am not sure if it is worth spending too much time investigating this. It is a minor annoyance and probably has always been there. Maybe the community Radio firmware will fix it [emoji3]

Sent from my Pixel 3a using Tapatalk

[mrw]

The network icon colours changed at some point to give less info. Now blue no longer exists. There is no difference now between a WiFi or LMS issue.

If you want to restore blueness, and you're feeling brave, the attached patch does it. (For me, anyway). E&OE.

Upload the attached patch file to your Radio, into the root user's home directory. Name it icon.patch

Then save the existing QVGAbaseSkinApplet.lua script, e.g. to QVGAbaseSkinApplet.lua.original, and apply the patch. Then check your work, and restart the Radio. You can restore the saved version back if things go wrong.

Here's a screen dump of the process as I undertook it:

# cd /usr/share/jive/applets/QVGAbaseSkin
# cp -p QVGAbaseSkinApplet.lua QVGAbaseSkinApplet.lua.original
# patch <~/icon.patch

[color=blue]Check your handiwork:[/color]

# diff -u QVGAbaseSkinApplet.lua.original QVGAbaseSkinApplet.lua 
--- QVGAbaseSkinApplet.lua.original
+++ QVGAbaseSkinApplet.lua
@@ -2385,10 +2385,10 @@
 		img = _loadImage(self, "Icons/icon_wireless_disabled.png"),
 	})
 	s.button_wireless_SERVERERROR = _uses(s._button_wireless, {
-		img = _loadImage(self, "Icons/icon_wireless_disabled.png"),
+		--img = _loadImage(self, "Icons/icon_wireless_disabled.png"),
 		-- In 7.6, decision is to go with only red colored icons for both network and server errors
 		-- leaving the old path here in case we decide that was a mistake
-		--img = _loadImage(self, "Icons/icon_wireless_cantconnect.png"),
+		img = _loadImage(self, "Icons/icon_wireless_cantconnect.png"),
 	})
 	s.button_wireless_NONE = _uses(s._button_wireless, {
 		img = false,
@@ -2401,10 +2401,10 @@
 		img = _loadImage(self, "Icons/icon_ethernet_disabled.png"),
 	})
 	s.button_ethernet_SERVERERROR = _uses(s._button_wireless, {
-		img = _loadImage(self, "Icons/icon_ethernet_disabled.png"),
+		--img = _loadImage(self, "Icons/icon_ethernet_disabled.png"),
 		-- In 7.6, decision is to go with only red colored icons for both network and server errors
 		-- leaving the old path here in case we decide that was a mistake
-		-- img = _loadImage(self, "Icons/icon_ethernet_cantconnect.png"),
+		img = _loadImage(self, "Icons/icon_ethernet_cantconnect.png"),
 	})
 
 	-- time

[bpa]

The error seems to be a red herring. If I close the error message I can see the file system. Enabling debug for net.http is extremely verbose. I am not sure if it is worth spending too much time investigating this. It is a minor annoyance and probably has always been there.

Logging is always verbose - trick is to find a way to minimise the traffic.
The issue seems like a minor but annoying niggle & difficult to identify. If it is battery associated - yet another potential level of complexity, I have no idea of what special measures are taken with the battery.