Read the statement by Michael Teeuw here.
Calendar Module Error with Many Google Calendars
-
I appear to be running into an error when I add many calendars from a single Google account.
The links for up to six appear to work, but then when adding more I get the following errors (edited for clarity and privacy)…
[2024-10-03 15:58:53.343] [ERROR] Calendar Error. Could not fetch calendar: https://calendar.google.com/calendar/ical/privatelink/basic.ics TypeError: fetch failed at node:internal/deps/undici/undici:13178:13 at process.processTicksAndRejections (node:internal/process/task_queues:95:5) { [cause]: ConnectTimeoutError: Connect Timeout Error at onConnectTimeout (node:internal/deps/undici/undici:2331:28) at node:internal/deps/undici/undici:2283:50 at Immediate._onImmediate (node:internal/deps/undici/undici:2315:13) at process.processImmediate (node:internal/timers:483:21) { code: 'UND_ERR_CONNECT_TIMEOUT' } } [2024-10-03 15:58:53.348] [ERROR] Calendar Error. Could not fetch calendar: https://calendar.google.com/calendar/ical/privatelink/basic.ics TypeError: fetch failed at node:internal/deps/undici/undici:13178:13 at process.processTicksAndRejections (node:internal/process/task_queues:95:5) { [cause]: ConnectTimeoutError: Connect Timeout Error at onConnectTimeout (node:internal/deps/undici/undici:2331:28) at node:internal/deps/undici/undici:2283:50 at Immediate._onImmediate (node:internal/deps/undici/undici:2315:13) at process.processImmediate (node:internal/timers:483:21) { code: 'UND_ERR_CONNECT_TIMEOUT' } }
Can anyone help me figure this out? When I get the errors, the calendar module appears to populate correctly and refresh for each calendar until it gets to the errors and then it clears that module and posts an error.
-
@moorevineyard timeout, network problem, it happens.
calendar should recover
-
please provide information about your mm-version and node/electron versions.
You find them in the logs, example:
[2024-10-03 23:10:32.636] [LOG] Starting MagicMirror: v2.30.0-develop ... [2024-10-03 23:10:33.352] [INFO] System information: ### SYSTEM: manufacturer: Microsoft; model: Docker Container; virtual: true ### OS: platform: linux; distro: Debian GNU/Linux; release: 12; arch: x64; kernel: 5.15.153.1-microsoft-standard-WSL2 ### VERSIONS: electron: undefined; used node: 22.9.0; installed node: 22.9.0; npm: 10.8.3; pm2: ### OTHER: timeZone: Europe/Berlin; ELECTRON_ENABLE_GPU: undefined
-
I’m open to that, but this seems like a very clear pattern, and it’s been hours without the missing calendars getting synched.
Also, it seems clear that the problem is related to the number of calendars I am trying to synch. It is always the first 6 that get added and the failure begins on the 7th. Everything after that point gets an error.
If I comment out one of the early calendars in the config file, I get the next 6 to synch and again the remaining ones get errors. Is there a timer that I can set to make them attempt a re-synch sooner than the fetchInterval in order to test this?
-
Is this all the relevant stuff?
> magicmirror@2.29.0 start > DISPLAY="${DISPLAY:=:0}" ./node_modules/.bin/electron js/electron.js [2024-10-03 17:27:24.356] [LOG] Starting MagicMirror: v2.29.0
[2024-10-03 17:27:28.036] [INFO] System information: ### SYSTEM: manufacturer: ; model: ; virtual: false ### OS: platform: linux; distro: Debian GNU/Linux; release: 12; arch: arm64; kernel: 6.6.31+rpt-rpi-v8 ### VERSIONS: electron: 31.6.0; used node: 20.17.0; installed node: 22.9.0; npm: 10.8.3; pm2: 5.4.2 ### OTHER: timeZone: America/New_York; ELECTRON_ENABLE_GPU: undefined
-
yes, thanks.
We had similar errors with older electron versions running with node v18, but you are using the newest mm release.
-
In case it matters, I also get a lot of these:
[8477:1003/172727.247634:ERROR:gbm_wrapper.cc(74)] Failed to get fd for plane.: No such file or directory (2) [8477:1003/172727.253029:ERROR:gbm_wrapper.cc(257)] Failed to export buffer to dma_buf: No such file or directory (2)
My searching on that makes it seem like it’s unrelated, but I’ve added the following into the mm.sh file and there’s no change:
export ELECTRON_DISABLE_GPU=1
-
Well, I did some testing and it looks like @sdetweil may have been correct.
I have a total of 9 calendars that I’m trying to synch. I noticed that the logs show execution of a “calendarfetcher”.
[2024-10-03 18:17:18.637] [LOG] Create new calendarfetcher for url: https:// ... .ics - Interval: 3600000
So it looks like the default interval is 3600000 (1 hr), but I thought the default fetchInterval was 300000 (5 min) based on the documentation. On top of that, I had a separate problem with the Pi that caused it to hang.
So that whole section of my logs looks like this…
[2024-10-03 18:17:18.637] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/first/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.758] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/second/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.763] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/third/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.766] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/fourth/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.770] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/fifth/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.774] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/sixth/basic.ics - Interval: 3600000 [2024-10-03 18:17:18.777] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/seventh/basic.ics - Interval: 60000 [2024-10-03 18:17:18.781] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/eighth/basic.ics - Interval: 60000 [2024-10-03 18:17:18.785] [LOG] Create new calendarfetcher for url: https://calendar.google.com/calendar/ical/ninth/basic.ics - Interval: 60000
I changed the fetchInterval for last three calendars to 1 min. At that point, those last three calendars are fetched, the screen gets updated, and the errors go away.
All seems good, right?
I did an additional test with the first 6 set to a 2 minute interval and the last three set to a 1 minute interval.
Again, the screen did not update properly for the first minute. At 1 minute, the last three calendars updated and the screen refreshed with all the info. After two minutes, all 9 calendars appeared to update without error and the screen stayed up-to-date.
So I think this is only a problem with the first run of the fetch function.
-
@moorevineyard 2 minute fetch is too frequent
it might not be enough time to get and parse
and DO you NEED to see the event you JUST added to the cal someplace else on the screen. when is the event? tomorrow, next week? not 2 minutes from now
-
I definitely agree that 2 minutes is too short for normal operations.
In my examples above, I was setting low times for troubleshooting purposes. I’m expecting to set the fetchInterval to about 24 hours once everything is set up the way I want. But this bug still causes lots of problems.
In particular:
For a new user that is frequently changing settings and styles, every change would require waiting a significant period of time before being able to check if it is acceptable. In my case, it means waiting a full day to see any calendar events again if I want to change the colors of one calendar.
Over time, events or calendars may be added by mistake. As far as I know, the normal way to refresh everything is to restart MagicMirror. Again, in this case, removing something that I do t want displayed means that I can’t see other events for a full day.
Is there another method of refreshing calendar data that would be less disruptive?