Joplin version: 1.2.6 (this issue already existed in some previous versions)
Platform: Linux
OS specifics:
It's expected the application opens normally without hanging the machine when using the Wayland protocol.
This only happens under Wayland protocol. Joplin opens immediately with normal cpu usage under Xorg.
Log from the terminal:
(electron) The default value of app.allowRendererProcessReuse is deprecated, it is currently "false". It will change to be "true" in Electron 9. For more information please check https://github.com/electron/electron/issues/18397
Log from developer tools:
DevTools failed to parse SourceMap: file:///tmp/.mount_joplinywn0Xg/resources/app.asar/gui/ResourceScreen.js.map
/tmp/.mount_joplinyw…r/lib/logger.js:126 23:12:28 Starting Clipper server on port 41184
/tmp/.mount_joplinyw…evelopment.js:11494 Warning: componentWillReceiveProps has been renamed, and is not recommended for use. See https://fb.me/react-async-component-lifecycle-hooks for details.
* Move data fetching code or side effects to componentDidUpdate.
* If you're updating state whenever props change, refactor your code to use memoization techniques or move it to static getDerivedStateFromProps. Learn more at: https://fb.me/react-derived-state
* Rename componentWillReceiveProps to UNSAFE_componentWillReceiveProps to suppress this warning in non-strict mode. In React 17.x, only the UNSAFE_ name will work. To rename all deprecated lifecycles to their new names, you can run `npx react-codemod rename-unsafe-lifecycles` in your project source folder.
Please update the following components: Connect(NoteTextViewerComponent), Connect(NoteToolbar), Connect(TagItemComponent), Connect(TagListComponent), Connect(ToolbarComponent), ReactAce
/tmp/.mount_joplinyw…evelopment.js:11494 Warning: componentWillUpdate has been renamed, and is not recommended for use. See https://fb.me/react-async-component-lifecycle-hooks for details.
* Move data fetching code or side effects to componentDidUpdate.
* Rename componentWillUpdate to UNSAFE_componentWillUpdate to suppress this warning in non-strict mode. In React 17.x, only the UNSAFE_ name will work. To rename all deprecated lifecycles to their new names, you can run `npx react-codemod rename-unsafe-lifecycles` in your project source folder.
Please update the following components: Connect(NoteTextViewerComponent), Connect(NoteToolbar), Connect(TagItemComponent), Connect(TagListComponent), Connect(ToolbarComponent)
3
/tmp/.mount_joplinyw…essageHandler.js:30 Got ipc-message: noteRenderComplete undefined
I updated the issue with logs from terminal and developer tools (I'm waiting for the log.txt been populated with information).
@laurent22, you removed the bug tag for some reason related to the lack of logs or because you do not consider a bug? Thank you.
this issue already existed in some previous versions
The GitHub tracker is only for actionable issues, and unfortunately while this bug might happen in your computer it's not possible to replicate it on our side. The log is just the usual warnings.
Does it happen with all the notes or just some of them?
And isn't it the same as this bug https://github.com/laurent22/joplin/issues/3114
Understand (about the tracker).
It's happening when I try to open the app. I'm aware of the issue with the kernel 5.5 but I missed the references to the high cpu usage and I decided to open this issue.
Can be related. I'll try later with a lower version of the kernel on another machine to be sure if is related or not.
I'll leave a message then. Thank you
I tested on another machine with kernel 4.19 and the result was the same: high processor usage when the application is opened.
(For me) Solved with →
Joplin 1.0.224 (prod, win32)
Revision: 1899d866 (master)
For me, the problem persists in the version 1.0.224.
The issue persists in the version 1.0.227 and kernel 5.4, 5.5, 5.6 and 5.7.
Note: Related with Gnome Shell and not with kernel?
@laurent22, I just made some progress to debug this issue.
This only happens under Wayland protocol. Joplin opens immediately with normal cpu usage under Xorg.
Please tell me if I can help providing more information.
Ive been stuck on 1.0.179 on fedora 32 for this very reason (ie I get the same electron message and then nothing).
I havent tried reverting the kernel because:
I keep trying every release (and some pre-releases) to no avail :(
(FYI I am using appimage home & config directories to keep attempts with other releases separated from my 1.0.179 data so they should be running clean)
I updated the issue with the Fedora 32.
@isaiahfrantz, can you test with Xorg and tell me if the issue persists?
I heard that the kernel issue had been resolved?
I tested with the old 4.19 kernel and I suspect this issue is not related with the kernel.
The issue persists in the version 1.0.233.
The issue persists in the version 1.0.242.
I have been following these types of threads since 1.0.207 (the last version that worked for me)
I've just discovered a fix for my situation, maybe it will help others. While trying to find some smoking gun I came across the fact that I could not update nodejs.
`Error:
Problem: cannot install both nodejs-2:14.11.0-1nodesource.x86_64 and nodejs-1:10.19.0-2.module_el8.1.0+296+bef51246.x86_64
Well that was short lived. It hangs again now on launch...I really don't know why.
I even installed 1.1.4 and still hangs on launch...
Since the upgrade to 1.1.4 I can launch the AppImage from terminal and it will open without hanging. At least there is a work around...
@JCTechSol, I have been using the terminal method but sometimes I have to cancel the process and start again to work.
I did some testing with Debian Unstable with a newer version of xwayland and it seems to have worked well.
Still, I can't confirm that it's due to xwayland because I ran out of time to continue my tests.
Today I wasn't able to open Joplin from terminal. It froze just like with the app shortcut. After repeated attempts and combinations I was able to get it to launch after I tried lunching an old version. It presents an error message about the database being newer then expected. After I press OK and dismiss that, I can then open the appimage of 1.1.4 successfully. I'm not sure why that combination works.
There is some more oddities with this workaround as well. I've discovered that if I try lunching the 1.0.241.AppImage it doesn't work, in fact it will crash the computer. But if I try lunching the 1.0.207.AppImage the database error successfully pops up saying that the database is newer then expected etc. Then I can successfully lunch the 1.1.4.Applmage. It seems like such a random piece of information but I can faithfully reproduce this behavior every time so I think its significant.
Hi, I am on Debian 10 and use the .deb from github releases. Same issue: Joplin freezes the system on Wayland. This is very counter-productive :( I lost some unsaved work and computations.
Update: The issue persists in the version 1.2.6.
I can also confirm the same behavior is present in 1.2.6. Also the same workaround works. Launching the 1.0.207 appImage, click ok on the error message, then launching the 1.2.6 appimage
I haven't be able to do a lot of testing but it appears 1.3 won't run at all. Thankfully I am still able to run 1.2.6 with the workaround as it looks like no modification\upgrade has happen on the database. (Launching the 1.0.207 appImage, click ok on the error message, then launching the 1.2.6 appimage)
Most helpful comment
The issue persists in the version 1.0.233.