Not really. The & is there to force Chromium running in the background. This way Chrome will no longer be "attached" to the terminal that instantiated it.
Probably running chromium from a script has a similar effect to running it explicitly using bash.
Hi @Colin and @Andrei ,
Thank you both for taking the time to help me.
The system has now been running for over 3 and a half hours and it's nearly midnight.
I'll leave it running, and see what has happened when I come down in the morning, which is usually when I find it has failed.
There are still some error messages but hopefully now as they're not being buffered they will disappear into the ether.
The system parameters are stable, cpu utilisation at about 2-3% and memory in use is also stable at under 60% used, despite this message appearing about 4 or 5 times a minuit from the stderr output on the exec node.
We shall see.......
Cheers.
Paul.
1 Like
Quite interesting to learn that Chrome generate so many error messages. Maybe they could be disposed to /dev/null if that helps. I just donĀ“t remember how to do that. I always get confused with the syntax.
Hi @Andrei,
I'm hoping that they are not now being buffered and will disappear into some bit bucket!
I'll know in the morning and will update.
I would like to get to the bottom of this though, and have posted on the google chromium help forum.
Try this small flow to see the errors.
[{"id":"3665f925.450a76","type":"exec","z":"e5593c90.29dda","command":"bash -c \"DISPLAY=:0 chromium-browser --noerrdialogs --disable-infobars --kiosk http://localhost:1880/ui -start-fullscreen\"","addpay":false,"append":"","useSpawn":"true","timer":"","oldrc":false,"name":"Kiosk Display","x":360,"y":80,"wires":[["c3c8d968.9e3898"],["9bc4ab56.ff2458"],["ae8236a8.de0d08"]]},{"id":"4ba01b75.94a9e4","type":"inject","z":"e5593c90.29dda","name":"","props":[{"p":"payload"},{"p":"topic","vt":"str"}],"repeat":"","crontab":"","once":true,"onceDelay":"1","topic":"","payload":"","payloadType":"date","x":170,"y":80,"wires":[["3665f925.450a76"]]},{"id":"c3c8d968.9e3898","type":"debug","z":"e5593c90.29dda","name":"stdout","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","statusVal":"","statusType":"auto","x":600,"y":40,"wires":[]},{"id":"9bc4ab56.ff2458","type":"debug","z":"e5593c90.29dda","name":"stderr","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","statusVal":"","statusType":"auto","x":600,"y":80,"wires":[]},{"id":"ae8236a8.de0d08","type":"debug","z":"e5593c90.29dda","name":"ReturnCode","active":true,"tosidebar":true,"console":false,"tostatus":false,"complete":"true","targetType":"full","statusVal":"","statusType":"auto","x":620,"y":120,"wires":[]}]
To really see the errors take off add this parameter "enable-logging --v=1"
Cheers.
Paul.
That should be --enable-logging. Without the -- it will try to open the url enable-logging. What does the --v=1 do?
Good morning @Colin & @Andrei ,
All working fine this morning and its now been running for 13 1/2 hours.
@Colin the --v=1 is verbose mode I believe.
I can see what you are saying about --enable-logging and you are correct,
See this link https://www.chromium.org/for-testers/enable-logging
And here is the current list of chromium switches Switches 
Just need to find the switch to suppress all error messages.
I will get to the bottom of this.
I didn't realise I had so much time to waste 
@Colin & @Andrei where in the world are you?
Cheers.
Paul.
Nice, good news.
Keep us updated. I suspect I faced this issue in the past but wrongly assumed it was related to Chromium automatic updates.
I will setup a new raspberry for a testing from scratch using the same configuration as you. Just need to buy a new SD Card. It may take a few days to have it tested.
I am hidden in a cave in Brazil waiting for the pandemic to slow down. Of course having a great time with Node-RED. 
Mid Wales, UK. Where are you?
I believe I have solved the problem by using the switch --log-level
It sets the minimum log level.
Valid values are from 0 to 3: INFO = 0, WARNING = 1, LOG_ERROR = 2, LOG_FATAL = 3
So I used it like this --log-level=3 and all error messages have stopped!
So the cmd line is now DISPLAY=:0 chromium-browser --noerrdialogs --disable-infobars --kiosk --start-fullscreen --log-level=3 http://localhost:1880/ui
I'm in Norwich, Norfolk, UK.
Thank you both for your help.
Cheers.
Paul.
1 Like
You have stopped it logging the errors, but they are still happening.
You are correct. Did you replicate the flow and see if you had the errors also?
It may be that there were always errors but no one has seen them.
I just finished to setup a new Pi. I confirm that I got the very same errors as you. I always see from 6 to 8 errors right after the flow is deployed. I need to let the Pi running a larger period of time to see if more errors appears. I did not take the time to search the meaning of those errors though.
[975:1027:0305/201703.651634:ERROR:bus.cc(393)] Failed to connect to the bus: Could not parse server address: Unknown address type (examples of valid types are "tcp" and on UNIX "unix")
[975:1126:0305/201704.914876:ERROR:object_proxy.cc(621)] Failed to call method: org.freedesktop.DBus.Properties.Get: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files
[975:1126:0305/201704.915777:ERROR:object_proxy.cc(621)] Failed to call method: org.freedesktop.UPower.GetDisplayDevice: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files[975:1126:0305/201704.917210:ERROR:object_proxy.cc(621)] Failed to call method: org.freedesktop.UPower.EnumerateDevices: object_path= /org/freedesktop/UPower: org.freedesktop.DBus.Error.ServiceUnknown: The name org.freedesktop.UPower was not provided by any .service files
[1020:1020:0305/201705.160159:ERROR:gl_surface_presentation_helper.cc(259)] GetVSyncParametersIfAvailable() failed for 1 times!
Hi @Andrei,
Yes those are the same errors that I was seeing.
I think I'm happy accepting that they are there and just not seeing them.
At least chromium on the Pi is not failing anymore.
What a huge list of switches chromium has!
log-level valid values are from 0 to 3: INFO = 0, WARNING = 1, LOG_ERROR = 2, LOG_FATAL = 3
The default log-level is 0 which means the messages are not necessarily errors. I didn't systematically go through the different values to see if they were INFO, WARNING or worse.
As I have shown earlier my cmd line in the exec node is now DISPLAY=:0 chromium-browser --noerrdialogs --disable-infobars --kiosk --start-fullscreen --log-level=3 http://localhost:1880/ui
And that works for me as I don't have the time or inclination to pursue the messages any further.
Interesting though, will you look deeper?
Cheers.
Paul.
Hi Paul, I am starting the think that those errors are not critical at all. First off, they are not increasing. I am monitoring with HTOP and donĀ“t see any performance issues. Memory and CPU utilization are low. No crashes observed. I would be concerned if those errors were due to my flows or Node-RED (or some misbehaving node) or Raspian config, which certainly is not the case. Those errors are Chromium specific. I will not dive deeper in the investigation unless some crash happens in my setup. Indeed the amount of configurable flags in Chrome is huge which makes troubleshooting even harder. Anyway it was good to track this thread as I have learnt a new way to trigger the kiosk mode.
Hi @Andrei,
I am always learning.
The nice thing about triggering kiosk mode this way, is that it can be turned on and off by enabling or disabling the exec node, and then doing a deploy from the development pc.
Nice chatting, I'm sure our paths will cross again.
Cheers.
Paul.
This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.
