@fribse Nope, just removed it and booted the mirror, still seeing those…

@sdetweil Glad to show you something new :rolling_on_the_floor_laughing:
I wonder if it’s the

gpu_mem=128

in boot/config.txt
?

Well, after doing a reinstall, module for module, it’s working with everything except that the Screencast module interferes with the remote module for the Home Assistant Integration (which uses the remote module).

So a success for now, and then I hope the integration to HA will get working.
Thans @sdetweil and @bugsounet !

@karsten13 I would normally agreee, but this is only to satisfy weird demands on SSL, not for any form of security and the mirror is not externally available.

@fribse Ok, so the black screen has nothing to do with the error:

0|MagicMirror  | [1554:0213/183501.650842:ERROR:viz_main_impl.cc(161)] Exiting GPU process due to errors during initialization
0|MagicMirror  | [1601:0213/183503.081619:ERROR:viz_main_impl.cc(161)] Exiting GPU process due to errors during initialization
0|MagicMirror  | [1621:0213/183503.548717:ERROR:sandbox_linux.cc(376)] InitializeSandbox() called with multiple threads in process gpu-process.

As that is still shown when I start the software…

@bugsounet Ha ha ha, no, I’m not blaiming anybody except myself. :beaming_face_with_smiling_eyes:
How do I
“So: Just use node v14.x”

@sdetweil I just did a reinstall, and now it displays stuff on the screen.
I do see this in the logs repeated over and over (as I said, fresh install):

0|MagicMirror  | [6213:0213/182653.081807:ERROR:socket_posix.cc(96)] CreatePlatformSocket() failed: Adressefamilien er ikke understøttet af protokollen (97)

Now I’ll try to add modules from my old config to see if it keeps working.

@sdetweil Can I attempt a reinstall by just renaming the MagicMirror folder?
I’ve also lost my most appreciated module, MMM-MQTTBridge, so I would love to just go back to the 2.17.1 if possible at all and the node.js and npm to be on an older (working) version :-(

@sdetweil I used your script (of course :smiling_face_with_halo: )
It’s a Raspberry Pi 3 Model B Rev 1.2
According to /sys/firmware/devicetree/base/model

node -v
v16.14.0
npm -v
8.3.1

I’ve now tried to only load the clock module, same result.

No message in the debug console, except the happy ones.

Btw. release shows:

/etc/*release
PRETTY_NAME=“Raspbian GNU/Linux 10 (buster)”
NAME=“Raspbian GNU/Linux”
VERSION_ID=“10”
VERSION=“10 (buster)”
VERSION_CODENAME=buster
ID=raspbian
ID_LIKE=debian
HOME_URL=“http://www.raspbian.org/”
SUPPORT_URL=“http://www.raspbian.org/RaspbianForums”
BUG_REPORT_URL=“http://www.raspbian.org/RaspbianBugs”

Hi all

One more to the list of people having problems after 2.18.
I’m running a RPi to show the magicmirror.
I can see on ‘http://mirror:8080’ that it renders the page.
But nothing is shown on the monitor
By now in my debugging I only have these modules activated:
“compliments”
“alert”
“updatenotification”
“clock”
“calendar”
“weather”
“newsfeed”
As far as I know, all core modules.
I see this in the logfile:

0|MagicMir | [974:0213/131905.747644:ERROR:viz_main_impl.cc(161)] Exiting GPU process due to errors during initialization
0|MagicMir | [1014:0213/131907.005140:ERROR:viz_main_impl.cc(161)] Exiting GPU process due to errors during initialization
0|MagicMir | [1036:0213/131907.699933:ERROR:sandbox_linux.cc(376)] InitializeSandbox() called with multiple threads in process gpu-process.

So a LetsEncrypt wildcard certificate could help here (if you have a domain that is). But does MM even support running with a cert out of the box, or do you need to recode stuff?

So I see this every morning:

0|MagicMirror  | [13.02.2022 12:34.28.850] [ERROR]
0|MagicMirror  | Calendar Error. Could not fetch calendar:  http://p01-calendarws.icloud.com/ca/subscribe/1/IsLh6fUURb4n_nA6vhWYUmlEe98rz2BN8BhioUPzrC2HVohwJIUCbGuGeTwczZWvXjTqU7vicBATXHojb1p5fn8ub3RCBUwKfHm3jXJDzFE Error: Service Unavailable
0|MagicMirror  |     at NodeHelper.checkFetchStatus (/home/pi/MagicMirror/js/node_helper.js:121:9)
0|MagicMirror  |     at processTicksAndRejections (node:internal/process/task_queues:96:5)

Which gives this result in the mirror:

Later in the day, the calendar days are fetched, and then updated on the screen.
Why does the module discard the existing days when it can’t get the new download?
My guess is that the calendar is somehow throttled, and thus rejects requests when there are too many, so how do I make the calendar module keep the old data until it can get new data?