Jump into Jammy Kubuntu 22.04 LTS

[dejjanku]

Did anyone had luck making nvidia drivers working with this new version ? 510 or ... ?

[jlittle]

The daily jammy iso today fails to isoboot for me, looping on

stdin: Not a typewriter

I forget why, but I've got the lubuntu daily iso too, and it has the same problem.

[MeMyself]

Well..... I've run into my first odd glitch on Jammy. Updated via Proposed, which I have been doing somewhat consistently without issue, The system has been and is running great.

The glitch I have run into seems to only effect Software Sources:

My password works perfectly for everything - except Software Sources, where I get the error shown in the attached image.

Steps to the Error:

Opened Discover
Settings Tab
Click Software Sources
Enter Password
---Error---

Opened Muon
Settings Menu
Configure Software Sources
Enter Password
---Error---

***If I select a Repository listed on the Settings page of Discover (either checking or unchecking the applicable box), the password works perfectly fine and processes the action

[Snowhog]

Same here. It'll get fixed.

[NoWorries]

Same here. It'll get fixed.

I also found this with Muon and I got around it by starting Muon with sudo and it then works for Configure Software Sources.

There is another more serious problem that I have and that is with Kdenlive on Wayland. With Wayland, after I do a render, I then try to run the rendered file with VLC and it fails to start. Even if I try VLC using the command line, that also fails. I also find that I cannot shutdown and I have to make sure that I have Konsole open so that I can do a

shutdown -h now

On X11 I do not have this problem after all the Plasma and Frameworks updates today, Kdenlive now works as it should. It renders OK and VLC works OK and I can logout. I hope the Wayland problem is easily fixed. My current work-around is to use X11 for all video editing with Kdenlive.

[oshunluvr]

I remember back while I was using Ubuntu 20.04 I tried to use "sudo do-release-upgrade" and it would not upgrade to 20.10 or 21.04 because they were not LTS release.
What I'm confused about is if I take my 21.10 system and do "sudo do-release-upgrade" after 22.04 LTS is released will it upgrade me to 22.04 LTS and keep me on the LTS?

OK, now I understand. The answer is the same. What you experienced before is because by default LTS releases only upgrade to LTS releases, whereas regular non-LTS release will upgrade to the next release regardless. To modify this behavior, you can edit /etc/update-manager/release-upgrades. In that file you will find the bottom line will be;

Prompt=normal

If you were using an LTS release, the default is;

Prompt=lts

I'm not sure what happens when you upgrade to an LTS release from non-LTS in this instance. Does the update change the release-updates entry or not? I guess we'll find out soon.

[pivert]

Firefox (default as Snap package) just loops on requesting to be the default browser on every start. Annoying.
Also, I experienced a problem with EID card reader. When plugging in, the default green led turns off after 5-10s... which is not normal. And the device does not work. I need to investigate.

[claydoh]

Firefox (default as Snap package) just loops on requesting to be the default browser on every start. Annoying.

This actually has little to do with it being a Snap, or even an anomaly of running a pre-release OS. it can happen on the deb version in 20.04 (on Neon) and should be fixed in FF 98

[jlittle]

Firefox (default as Snap package) just loops on requesting to be the default browser on every start. Annoying.

To remove that annoyance, use about:config to set browser.shell.didSkipDefaultBrowserCheckOnFirstRun to true.

As Claydoh reports, a fix is in the pipeline. This nuisance has been there for years, and they've always refuse to divulge exactly what is being checked. So, I'm sceptical...

[rec9140]

So where is the best place to send feedback that will garner attention on issues

The below is some feedback of my testing this AM... for the curious... like above, I am looking for the BEST STOP for it to be sent... Launchpad??

Testing 22.04 to see how things shake out. FULLY AWARE this is alpha/beta at best testing editions from the daily builds....

Using a VM for now, no spare physical hardware that can be used

1) Updated to all current packages in repos (have NOT added any repos, just whats there by default)
2) purged firefox snap and snapd - good riddance!

Things seems to be fine up till here!

3) Switched to LightDM as I REQUIRE XDMCP... and I think this is where things and wheels on the bus fall apart and off!

a) Using the BUILT IN RDP server of the hypervisor, it started to show kinks in the armor... reboot to apply LDM setup, get the greeter, OK.. login.... wait and wait and wait... black screen... hmmmm... ok disconect reconnect.. there is the desktop...

b) The desktop is there but you can't click on anything.. you move the mouse and see it move, but clicking on anything, NOTHING on the RDP via KRDC (20.04) nothing.. BUT AHAHA! Its showing on the VM screen!!!!!! (If I go look on the box in the other room) and the input where I clicked to bring up the menu, and typed into syanpitc some gibberish there it was ... hmmmmm...

c) Try an direct XDMCP connection. xephyr to the box Xephyr -retro :1 -query xxx.xxx.xxx.xxx

i).. ok brings up the greeter... ok... enter password.. "Failed session" enter again, making sure I type correctly , it takes and attempts to start a session, nope, not going to happen. returns back to the greeter screen

ii) SSH in to the 22.04, do a Xephyr -retro :1 -query 127.0.0.1 same thing
iii) SSH in start a program with X forwarding... firefox --no-remote... first start up took forever (had not started since I did the purge, install via DEB (apt CLI), seems to start within reason after...

So hmmmm.. seems LightDM is the break down here.. AGAIN LET ME REITERATE, I get it, its alpha/beta testing period... I was really hoping not to see these issues in re LightDM, but I had suspicions with this whole wayturkey mess... I had great success with 20.04 at this point on a generic OEM (motile 142 laptop) I got a deal on... just a little bummed to see these issues crop in as I suspected. ....

I am going to rebuild the VM from pristine to the point before I add LDM, I had not rebooted after all the updates... so I want to see if that breaks the RDP server of the hypervisor..or its LightDM, I pretty much have my suspicions, sadly.

NOTE: I am on X11, and will be remaining on X11, period. Thats not a debate for here and now, and even later its pointless with me. That red line is set.
Note 2:This is a pure Linux environment, we have no winslopper. Have not for 22+ years.

Note 3: I am testing in a VM, and and its not on some Ryzen box, but it does the job...I've tested everything from 16.04 to 20.04 on here for years. with no issues other than what my fat fingers created! VM carcasses all over!

Any way where is the best spot for this feedback after I nail it down to get to some place where it will be addressed. Thanks in advance.

Comments on the issues and resolutions or other insight, welcome as well. ( Happy to debate as adults on X11 v. wayturkey, same in re snapdud, at at different time/place etc...but not right now)

[Snowhog]

The short answer is you should have/should registered/register to be a Kubuntu Tester: https://wiki.ubuntu.com/KubuntuTesters

This really is the only effective way to get pre-release issues properly identified and reported.

[NoWorries]

On my laptop I have three partitions with Windows 10, Jammy and Impish. The final fdisk -l partitions are:

Disk /dev/nvme0n1: 953.87 GiB, 1024209543168 bytes, 2000409264 sectors
Disk model: Samsung SSD 970 PRO 1TB
Units: sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disklabel type: gpt
Disk identifier: 9D947ABE-5180-4015-871F-25FDC5F829D0

Device Start End Sectors Size Type
/dev/nvme0n1p1 2048 534527 532480 260M EFI System
/dev/nvme0n1p2 534528 567295 32768 16M Microsoft reserved
/dev/nvme0n1p3 567296 230363555 229796260 109.6G Microsoft basic data
/dev/nvme0n1p4 250370048 252223487 1853440 905M Windows recovery environment
/dev/nvme0n1p5 252223488 287379455 35155968 16.8G Linux filesystem
/dev/nvme0n1p6 287379456 1110806527 823427072 392.6G Linux filesystem
/dev/nvme0n1p7 1110808576 1112762367 1953792 954M Linux swap
/dev/nvme0n1p8 1112762368 1147918335 35155968 16.8G Linux filesystem
/dev/nvme0n1p9 1147918336 2000408570 852490235 406.5G Linux filesystem

I found that Kdenlive worked well on Impish Wayland with all other related applications. However with Jammy Wayland, kdenlive hung the desktop and I could not Logout.

I therefore decided to do a fresh installation of Jammy and use the default Wayland packages. This is when all hell broke loose. The most noticeable problem on the re-installation was that it only booted Jammy. No Windows or Imprish boot option was visible!!!!

This is the first time I have encountered this problem and possible solutions relate to the EFI partition being too small - which it is not in my case.

Does anyone know how to solve this problem. I am sure that if I re-install Imprish this will fix it, but this will be the last resort for me.

[jlittle]

You could get the grub in jammy to boot to impish and windows, by writing a custom.cfg and dropping it into /boot/grub.

Here's one I've used in the past:

set default="0"

insmod all_video
insmod part_gpt
insmod btrfs
set timeout=30
set menu_color_normal=white/black
set menu_color_highlight=black/light-gray

menuentry 'Kubuntu from custom' --class ubuntu --class gnu-linux --class gnu --class os {
search --no-floppy --label --set=root main

linux /boot/vmlinuz root=LABEL=main ro 
initrd /boot/initrd.img
}

menuentry 'Windows' --class windows --class os {
insmod part_gpt
insmod fat
#search --no-floppy --fs-uuid --set=root 5EBB-BCE4
search --no-floppy --label --set=root SYSTEM
chainloader /EFI/Microsoft/Boot/bootmgfw.efi
}

You'd have to change the label names, or change the search statements to use --fs-uuid, the linux line to use root=UUID=, and supply the UUIDs of the Windows and impish partitions. After putting custom.cfg in /boot/grub, there's no need to run update-grub. (If impish uses btrfs you'd prepend "/@/" to the vmlinuz and initrd.img file names, and the phrase "rootflags=subvol=@" to the linux line.) It might pay to check in the ESP that the Microsoft efi file has the right name.

[NoWorries]

After updates to Jammy this morning, I was once again faced with Jammy being my only OS. Windows 10 and Kubuntu 21.10 were not listed in the boot options.

I found a very simple solution to this problem at: https://arcolinuxforum.com/viewtopic.php?t=2668

The command sudo update-grub does not run os-prober which actually lists to missing OS's. So to get os-prober to be run by grub, I edited:
/etc/default/grub
and added the line:

# Check for other operating systems
GRUB_DISABLE_OS_PROBER=false

This was then followed by

sudo update-grub

All OS's were now found after this step and and all could be accessed.

[verndog]

You could get the grub in jammy to boot to impish and windows, by writing a custom.cfg and dropping it into /boot/grub

I've been using a custom 'grub.cfg' for years without issue. Using 'labels' makes reading easier:

insmod gfxterm
set gfxmode=640x800
set timeout=30

set menu_color_normal=cyan/blue
set menu_color_highlight=white/blue

menuentry 'Windows' --class windows --class os{
insmod part_gpt
insmod fat
search --no-floppy --fs-uuid --set=root 2787-600D
chainloader /EFI/Microsoft/Boot/bootmgfw.efi
}
menuentry 'ubuntu' --class ubuntu --class gnu-linux --class gnu --class os{
insmod all_video
insmod gzio
insmod ext2
search --no-floppy --set=root --label "ubuntu"
linux /boot/vmlinuz root=LABEL=ubuntu ro quiet splash pci=noaer vt.handoff=7
initrd /boot/initrd.img
}
menuentry 'neon' --class ubuntu --class gnu-linux --class gnu --class os{
insmod all_video
insmod gzio
insmod ext2
search --no-floppy --set=root --label "neon"
linux /boot/vmlinuz root=LABEL=neon rw quiet loglevel=3 nowatchdog
initrd /boot/initrd.img
}
menuentry 'kde' --class ubuntu --class gnu-linux --class gnu --class os{
insmod all_video
insmod gzio
insmod ext2
search --no-floppy --set=root --label "kde"
linux /boot/vmlinuz-linux root=LABEL=kde rw quiet loglevel=3 nowatchdog
initrd /boot/intel-ucode.img /boot/initramfs-linux.img
}
menuentry 'xfce' --class ubuntu --class gnu-linux --class gnu --class os{
insmod all_video
insmod gzio
insmod ext2
search --no-floppy --set=root --label "xfce"
linux /boot/vmlinuz-linux root=LABEL=xfce rw quiet loglevel=3 nowatchdog
initrd /boot/intel-ucode.img /boot/initramfs-linux.img
}

menuentry 'fwsetup'{fwsetup}

submenu ISO {
menuentry "ISO" {
rmmod tpm
set isofile="/ISO/ubuntu.iso"
loopback loop (hd1,1)$isofile
linux (loop)/casper/vmlinuz boot=casper toram maybe-ubiquity iso-scan/filename=$isofile noprompt noswap noeject
initrd (loop)/casper/initrd
}
menuentry "clonezilla" {
gfxpayload=1024x768x16,1024x768
set isofile="/ISO/clonezilla-live-2.7.2-39-amd64.iso"
loopback loop (hd1,1)$isofile
linux (loop)/live/vmlinuz nomodeset boot=live live-config edd=on ocs_live_run=\"ocs-live-general\" ocs_live_extra_param=\"\" keyboard-layouts=\"us\" ocs_live_batch=\"no\" locales=\"en_US.UTF-8\" ip=frommedia toram=filesystem.squashfs findiso=$isofile
initrd (loop)/live/initrd.img
}
}

I replace my 'grub.cfg' whenever it gets overwritten