Getting strange error after upgrade with Electron

junkmale

updated to node 19.6.0

Still having the following error:

/home/pi/.pm2/logs/mm-error.log last 50 lines:
0|mm | Gdk-Message: 12:52:54.219: electron: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.
0|mm |
0|mm | [08.02.2023 12:53.25.900] [WARN] You’re using a full whitelist configuration to allow for all IPs
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to open kms_swrast (search paths /usr/lib/arm-linux-gnueabihf/dri:$${ORIGIN}/dri:/usr/lib/dri)
0|mm | failed to load driver: kms_swrast
0|mm | MESA-LOADER: failed to open swrast (search paths /usr/lib/arm-linux-gnueabihf/dri:$${ORIGIN}/dri:/usr/lib/dri)
0|mm | failed to load swrast driver

I then did a sudo apt-get update then a sudo apt-get upgrade; it upgraded a couple of xserver files. Cleared the PM2 log and rebooted. Same errors as above.

sdetweil

@junkmale ok, after upgrading node. then u need to refresh mm install

do

pm2 stop all
cd ~/MagicMirror
rm package-lock.json
rm -rf node_modules
npm install --omit=dev --no-audit --no-fund
pm2 start all

junkmale

That didn’t seem to do it. After a pm2 flush, it shows

/home/pi/.pm2/logs/mm-error.log last 50 lines:
0|mm | [1176:0208/151743.603505:ERROR:connection.cc(46)] X connection error received.
0|mm | Gdk-Message: 15:17:43.634: electron: Fatal IO error 11 (Resource temporarily unavailable) on X server :0.
0|mm |
0|mm | [08.02.2023 15:18.16.665] [WARN] You’re using a full whitelist configuration to allow for all IPs
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to retrieve device information
0|mm | MESA-LOADER: failed to open kms_swrast (search paths /usr/lib/arm-linux-gnueabihf/dri:$${ORIGIN}/dri:/usr/lib/dri)
0|mm | failed to load driver: kms_swrast
0|mm | MESA-LOADER: failed to open swrast (search paths /usr/lib/arm-linux-gnueabihf/dri:$${ORIGIN}/dri:/usr/lib/dri)
0|mm | failed to load swrast driver
0|mm | [998:0208/151919.102171:ERROR:gl_surface_presentation_helper.cc(260)] GetVSyncParametersIfAvailable() failed for 1 times!
0|mm | [998:0208/151919.105457:ERROR:gl_surface_presentation_helper.cc(260)] GetVSyncParametersIfAvailable() failed for 2 times!

sdetweil

@junkmale did u add that env setting to disable the gpu?

is this the 64bit os?
uname -a

junkmale

added the GPU setting to mm.sh; this is a Pi 3, ARMv71.

junkmale

@sdetweil Not sure if the last response took or not. GPU edit to mm.sh is complete, I did a hard boot afterwards. Response to uname command:

Linux MagicMirror 5.10.103-v7l+ #1529 SMP Tue Mar 8 12:24:00 GMT 2022 armv7l GNU/Linux

sdetweil

@junkmale but it still shows active in the pm2 output. did u create a different script for pm2, like in the root of the mm project?

my install script uses the installers version

junkmale

@sdetweil not that I know of.

pm2 show mm says the script path is /home/pi/mm.sh;
/home/pi/mm.sh did not have the GPU command in it.
I did change the one in ~/MagicMirror/installers, but since that doesn’t seem to be the right one,
I changed the one in /home/pi and pm2 restart mm - and now we have zero errors in the pm2 log.

One down.

However…MMM-RAIN-MAP still shows diddly-squat. And it’s still not showing in the right place.
npm config check comes back without errors. By the way, I thought I posted the question in the MMM-RAIN-MAP forum, but perhaps an error on my part.

junkmale

@sdetweil Wait! - I think I got it!

Thank you!

sdetweil

@junkmale yes, rainmap has a bug on memory I think. the author needs to fix it. open an issue there