How to fix CAPS LOCK syncing in NoMachine? - server

I have searched everywhere for a fix to the keyboard CAPS LOCK bug for NoMachine, but I cannot seem to find any.
The issue is that the CAPS LOCK is acting extra weird where it turns itself on and off at irregular times (even when not requested), not turning at all, or it may get stuck at one state.
I am using both of client/server on Mac OS High Sierra and Windows 10, and I tried connecting between Mac-Windows, Windows-Mac, and even Windows-Windows with another Windows 10 device, and the issue still persists. I assume this issue happens on Linux client/server as well.

There have been some problems, in the past, due to the inheritance of some keyboard code from X11, but, as far as I know, no such problems exist, in the current version. Are you using a reasonably recent version? The latest is 6.3.6.

Although the bug report that NoMachiner mentioned in a comment has been closed with the comment "Resolved in version 6.10.12 released on 2020-05-12.", this issue still seems to be present in version 6.11.2, although it doesn't happen every time.
The way I fix it is to give focus to any application other than NoMachine, toggle the caps lock and then return to NoMachine. It seems that for some reason, while NoMachine has focus toggling Capslock will always toggle it, but when NoMachine doesn't have focus, it will only toggle the capslock if the state has changed. Very strange, and very annoying.

I solved it by following these instructions on ArchLinux wiki (Map_scancodes_to_keycodes):
Created the file /etc/udev/hwdb.d/90-block-sleep.hwdb and added the following lines:
evdev:atkbd:dmi:bvn*:bvr*:bd*:svn*:pn*:pvr*
KEYBOARD_KEY_3a=leftctrl
Next run:
$ sudo systemd-hwdb update
$ sudo udevadm trigger
Maybe you need adapt the solution to your keyboard.

Disconnect noMachine
Toggle the Caps Lock outside of noMachine
Re-connect noMachine
This worked for me.

Related

How to fix VS Code integrated terminal choppy and laggy on macOS Big Sur?

Typing in the integrated terminal in VS Code on Big Sur is very annoyingly laggy and slow. This is the only application that experiences this, iTerm, etc. do not experience this issue. I tried disabling all plug-ins, themes, fonts, etc. to try to get to root cause.
Based on the thread here I was able to successfully mitigate the issue by running this command: codesign --remove-signature /Applications/Visual\ Studio\ Code.app/Contents/Frameworks/Code\ Helper\ \(Renderer\).app and restarting VS Code, the lag is completely gone and everything is back to normal.
If you trust VS Code developer then removing the signature should not be an issue but YMMV.
For me, I was super happy to find this solution, as it was driving me crazy.
Probably an update on Konstantin Popovskikh's answer that was deleted.
In settings.json, setting this:
"terminal.integrated.gpuAcceleration": "off",
Helped a lot for me, however it does have the side effect of not rendering underscores while typing - but for me, this is the lesser of two evils.
Note: You don't appear to have to reload the session you can just save the .json file and the terminal will update (you can visibly see it change slightly).

vim emulation extension in VScode causing problems

I've been using VS Code in "VIM mode" for years now without a problem. I'm on v1.46 now. I fired it up today and immediately couldn't type and other keys like escape failed to register. In the lower left corner, I'd intermittently see the following:
I went into extensions and disabled the Vim Emulator which now allow me to edit the file normally... without the vim functionality, of course.
I saw a bug filed through GitHub that looked like it might have addressed it, but it was closed due to lack of info provided.
Any ideas what might be going on or how I can get my vim functionality back?
UPDATE
Here's a bit more info:
I reinstalled and updated both VSCode and plugin - same behavior
Sometimes the whole thing failed from the get and other times (after restarting), it seemed to fail 1 key at a time - for example I'd be typing along and then all of a sudden hitting backspace produces an error message in the lower right corner to the effect of command 'extension.vim_backspace' not found - here's the closed issue on github - https://github.com/microsoft/vscode/issues/22365 for that specific message, but I also got errors with the escape and other keys
I also tried switching to NeoVim (on mac) which I thought was working, but eventually ended up going the same route (esc stopped working first typically)
I was having a similar problem with Vim in VS Code, and I noticed at the bottom-left when it reads --NORMAL-- or --VISUAL-- I typed the letter "i" and --INSERT-- appeared and allows me to type. I typed "i" because when I could type that is what I saw in the bottom-left corner of the screen.
Hope this is helpful.

How to track down source of weird code insertion

In VS Code 1.10.2 running on macOS Sierra (10.12.3), whenever I type an open paren ( ( ) followed by two spaces in quick succession, a period ( ( . ) is inserted in my code. If I type the spaces with a momentary delay in between, nothing happens (no period is inserted).
I recently reformatted my hard drive and reinstalled my OS, so I know I'm working with a very clean box.
I've got 4 plugins installed: Apex, ESLint, ForceCode and Visualforce. I tried disabling all of them and reloading my editor, but that didn't fix the issue.
Is this a bug, a configuration setting, or something else? It's getting really annoying, so I'd like to find a way to make this stop.
I found the answer. Apparently this is a new MacOS Sierra feature. It's supposed to be turned off by default, but apparently it was enabled, perhaps because I installed a clean copy rather than upgrading.
You can disable this feature by opening Settings > Keyboard > Text and unchecking the "Add period with double-space" option:
Why on earth anyone would have thought that this feature would be useful I cannot fathom. It doesn't make sense in the context of sentence (since it places a space before and after the period)... but there it is.

Ubuntu 16.04 suddenly starts booting into Kodibuntu or something like it

Last week, I saw a dropdown next to the login field for Ubuntu with a few "boot options" or something like that, one of them being Kodi. I was curious and wanted to see what it was so I picked Kodi and rebooted. Nothing happened, and I forgot about it.
Today I had to reboot for another reason and Ubuntu is now booting into a fullscreen version of kodi (Kodibuntu?) with seemingly no way to exit. I can't alt-tab, I can't open a terminal (neither ctrl-alt-t or f1/f4 work). I looked in the system settings and I see no way to revert it. I can see that my document's are still there using Kodi's file browser but can't access any of them. I really need this fixed today, please help.
Check out this documentation on kodibuntu for how to swap back
you can exit the full screen version of kodibuntu by following the steps outlined in the link above.
Sorry if this is unclear but a screenshot would be useful to know what screen you're on.

Bootloader countdown timer stuck at 4 seconds

I seem to have a problem for one the servers that i am handling, i have reinstalled CentOS for nth times, and formated everything, edited the grub.conf time out = 0, but all still the same, the grub timer wont move, always stuck.
Did you end up finding a solution to this? I just had a server with this exact issue right after a fresh install of CentOS 6.x. Even though the system seemed to be keeping correct time, replacing the CMOS battery resolved the issue for me. This was either due to a faulty battery, or the CMOS settings getting reset when I replaced the battery.
Here are the sites that led me to that solution:
https://serverfault.com/questions/42877/grub-hangs-after-x-2-seconds
http://www.tomshardware.com/answers/id-1809044/centos-grub-countdown-stops.html
A user on one of those sites said that clearing the "System Event Log" fixed the issue for them. Here are some guides on accessing and clearing the SEL/BMC log:
http://download.intel.com/support/motherboards/server/s7000fc4ur/sb/sel_viewer_ug_e1246101.pdf
http://www-947.ibm.com/support/entry/portal/docdisplay?lndocid=MIGR-59226
The issue could also be related to Grub's recordfail logic. More info here:
https://askubuntu.com/questions/202309/cannot-get-grub-menu-to-timeout-or-go-away?newreg=dc178d813fbb4f0d803644799bb89c5e
https://bugs.launchpad.net/ubuntu/+source/grub2/+bug/872244
Try adding this to /etc/default/grub
GRUB_TIMEOUT=10
GRUB_RECORDFAIL_TIMEOUT=$GRUB_TIMEOUT
Then run sudo update-grub
Support for GRUB_RECORDFAIL_TIMEOUT was added in the middle of the 12.04 cycle, starting from version 1.99-21ubuntu3.3
Check the date/time setup of your server. After I changed the CMOS battery, server's (HP Proliant) clock was 4 days behind and had same issue, the problem was fixed after setting it up correctly.