KLV-Airedale-beta25 with Debian Kernel 6.0.0-3-KLV SMP PREEMPT DYNAMIC

[rockedge]

KLV-Airedale-beta25
Kennel Linux Void Airedale Beta 25 625 MiB download size ->
KLV-Airedale-beta25
SHA1-MD5.txt

This is constructed from KLV-Airedale-beta24 and with a custom Debian kernel version 6.0.0-3 SMP PREEMPT DYNAMIC made with mention]fredx181[/mention]'s newest script that will create a Debian kernel for KLV.
Uses an initrd.gz based on the FirstRib skeleton initrd.gz configured by the kernel/module creation script. Firmware SFS is from a woof-CE kernel-kit built huge 6.0.0-1-KLV kernel.

Also included is wiak's generate boot stanzas script. fredx181's mksplash has a menu entry and the remaster script is installed.

With the addition of a xfce4 desktop tiling configuration. A tip from @wiak for configuring Tiling windows done with the Super (Windows) key and an up, down, left, right, home, pgup, pgdown or end key.

gxmessage in the default rootfs.

save2flash has been expanded to handle persistence on FAT32 and NTFS partitions.
The mksplash GUI utility to create splash notification banners.

The latest initrd.gz brings vFat and NTFS formatted partition support to KLV, allowing persistence on NTFS/vFat systems as well as the usual /ext2/ext3/ext4 /swap support.

• included htop

• created a symlink /root/spot with the target /home/spot

• Auto login cleaned up.

• includes also the logout logic provided by fredx181 gives true mulit-user support.

• replaced loop.cfg to the most recent modifications.

• Debian kernel 6.0.0-3-KLV SMP PREEMPT DYNAMIC is used.

• added gtkhash

Plus the other important improvements contributed by the KLV team.

Added @fredx181's save2flash utility that can be used when KLV is started in RAM2 mode to perform session saves on demand for persistence.
This is similar to PUPMODE13 in Puppy Linux.

fredx181's swap partition enable script includes the latest revisions and also added are the packages

• mime-add-1.0_0.noarch.xbps

• edit-sfs-1.0_0.noarch.xbps

KLV-Airedale-beta25 is able to load squash SFS packages on the fly or during the boot sequence.
Squash files and ISO images can be opened from the file manager and the read the contents.

• can run with rootfs, 01fimware and 00module SFS files as uncompressed directories.

• Also able to load compressed and/or decompressed directories,
  once a 2 digit prefix is added to the file name.

• It is possible to mix using compressed squash files and uncompressed directories.

Also can be downloaded from https://rockedge.org/kernels in ISO->Kennel_Linux->Airedale

[rockedge]

I would like to experiment with lz4 compression of the KLV squash files.

To follow in a new topic will be how to remaster using the "frugal psuedo full install" method and running KLV in this mode.

[geo_c]

So I have KLV installed on 4 on four separate laptops now, and two separate installs on a USB stick that are mirrored, all of them with the same collection of numbered upper_changes. One of the USB installs is KLV-airedaleDEV, and I'm using that one to drop new beta versions and test with all my current upper changes.

I have a question about the iso before I get started. I'm looking in the grub boot folder, and at the wd_grubconfig, and wondering if that is mainly there to boot from an iso, and if that's at all useful to me, or should I just go with my tried and true method of using my jackalpup grub2 and modifying the stanza?

And I'm also wondering if my current grub boot needs to even be changed. Will my current grub find the files after dropping in the new beta files? Seems like it should (as long as I don't change the name of the install folder) because it's only looking for the location of vmlinuz and initrd.gz.

Code: Select all

title Linux KLV-airedale20-2 (sdb2/KLV-airedale20-2-UUID)
  find --set-root uuid () 1b186e42-4f56-4cf3-bc41-863feab8ee43
  kernel /KLV-airedale20-2/vmlinuz w_bootfrom=UUID=1b186e42-4f56-4cf3-bc41-863feab8ee43=/KLV-airedale20-2 net.ifnames=0
  initrd /KLV-airedale20-2/initrd.gz  

title KLV-airedale20dev (sdb2/KLV-airedale20dev-UUID)
  find --set-root uuid () 1b186e42-4f56-4cf3-bc41-863feab8ee43
  kernel /KLV-airedale20dev/vmlinuz w_bootfrom=UUID=1b186e42-4f56-4cf3-bc41-863feab8ee43=/KLV-airedale20dev net.ifnames=0 
  initrd /KLV-airedale20dev/initrd.gz

And glad to see the forum is working like it's old self again.

[rockedge]

@geo_c When what you are doing is working then stick with it. No need to switch from how you have it setup.

the script is to make boot stanza's for frugal installs and is for those cases one doesn't know the UUID or LABEL's of the partitions yet. It is included for the convenience for experienced users and a guide for the new user.

Getting the server fixed distracted from getting beta25 squared away. Glad it's working. Pays to be nice to IT support techs

The boot folder does not need to be in the frugal installed file system/structure.

You're correct that /boot is for booting from ISO or a burned CD/DVD or USB drive.

Also @fredx181's cr-debkernel script will be included in the next updated model of beta25

[geo_c]

Thanks @rockedge

My grub boots are all working as expected. Everything seems to be running smoothly. The beta25 verbose boot has a lot more going on.

Everything seemed to be working so I updated all my installs, and the only snag I've hit so far is trying to run my LibreWolf appimage. I get a message that says:

Cannot mount AppImage, please check your FUSE s...
You still might be able to extract the contents of this AppImage if you run it with the appimage-extract option. So I checked this out: https://github.com/Appimage/AppImageKit/wiki/FUSE

It also mentions extracting the AppImage. I'm guessing this might be a permissions thing?

@mikewalsh's portable Ungoogled Chromium and Palemoon working fine.

[fredx181]

For some reason the "fuse" kernel module is not loaded, to fix: in terminal modprobe fuse and the Appimage should run OK then.

[rockedge]

@geo-C I checked on FUSE which is installed but I notice that fuse-overlayfs is not, so I'll add that if you can test this by installing

Code: Select all

xbps-install -Sy fuse-overlayfs

or using the OctoXBPS interface.

Please test this theory out, I am dealing with the host provider....again.....to try to find a solution to the websites running so slowly.

[geo_c]

I ran the xbps install @rockedge suggested, which was successful and rebooted for good measure, but that did not work, then I ran @fredx181 modprobe suggestion which did work,

So the appimage fired up and is running.

I don't know if it needed the combination or just the modprobe command.

I'll reboot and see and if works without running the command again...

Rebooted and it did not work. I had to run the modprobe command again. As a temporary fix I suppose I could put that command in a startup script.

[rockedge]

Did this occur with beta24 and the Debian kernel 6.0.1?

[geo_c]

Did this occur with beta24 and the Debian kernel 6.0.1?

I never upgraded. I've been running off beta20-2 this entire time. So I jumped from beta 20 straight to beta25.

I put the modprobe command in a script in the /Startup folder and that works.

[geo_c]

I should report some quirky behavior that I've been experiencing in beta20 and is still persistent in beta25.

First, about 50% of the time when the desktop boots it pops a rox window up showing the desktop background directory that the background is chosen from. I would tend to think that I just left the window open on last run, but I can say for certain now that this not the case.

Second, about 50% of the time when the destop boots, I get a rox panel with two directories in it, /home and /Apps or /applications. I normally just remove this panel, but it periodically comes back for no apparent reason. I checked my /root/.config/rox.sourceforge.net/Rox-filer/panels file, and there's nothing in there. Also nothing in the /Startup folder. I always chalked it up to something I might have done, but I can't find it. Maybe it's in ../Rox-filer/Options, but I don't see an option that would be starting one.

Just raising it to see if anyone else sees this kind of behavior.

[fredx181]

Did this occur with beta24 and the Debian kernel 6.0.1?

For me, on beta24, yes (just now upgrading to beta25).
Strange..., cannot explain, fuse is in the list of modprobe'd kernel modules in the init script (in initrd.gz) but at boot it's not loaded.
Simple solution as @geo_c wrote is to put modprobe fuse in a script in ~/Startup or, perhaps make it as a runit service.

[rockedge]

Just raising it to see if anyone else sees this kind of behavior.

I have also experienced the panel coming on. This has happened on and off for a while. Never have pinpointed it yet why. I think it has something to do with using an upper_changes from an earlier version but not really sure about this.

[williams2]

when the destop boots, I get a rox panel with two directories in it

If you create a new rox panel (panel as opposed to pinboard)
You can create as many panels as you like, but each panel should have a different name.
And You can create as many pinboards as you like, each pinboard should have a different name.

For help type rox --help

When a new panel is created, there will be no configuration
so the panel will be created with two icons
for $HOME and /usr/local/apps/
Rox will create a configuration file after you change the configuration for that panel or pinboard
if you do not change something, for example, moving an icon or deleting an icon or adding an icon etc etc etc
then there will be no config file in /root/.config/rox.sourceforge.net/ROX-Filer/
a config file in /root/.config/rox.sourceforge.net/ROX-Filer/ will be created after you change something.

There is a list of panel names in the file panels
Seeing the name of the panel that was mysteriously created might be usefull

If you are trying to use session files for rox,
that may be causing the panels to be created, and it might be causing rox to have more than 1 instance running.
Normally, there will be 1 instance of rox and each rox window is created by that 1 rox instance.

So you need to check that only 1 instance of rox is runing
maybe using ps or htop

[fredx181]

@rockedge beta 25 doesn't boot for me (does it for you or anyone ?), tried to figure out why and tested cr-debkernel and found that it downloads newest 'linux-image-amd64' from "experimental" branch (linux-image-6.1.0-0-amd64-unsigned_6.1~rc3-1~exp1_amd64.deb, I think that one is included in beta25) (experimental branch often hasn't all included that's required for a frugal install, I guess)
Changed cr-debkernel script: 2022-11-07, bug fix, filter out 'linux-image-....' .deb from 'experimental' branch
https://forum.puppylinux.com/viewtopic. ... 909#p69909
Ran new cr-debkernel (it downloaded linux-image-6.0.0-3-amd64-unsigned_6.0.7-1_amd64.deb), swapped the 00modules .sfs and vmlinuz and initrd.gz in my beta25 install and it booted OK.
Sorry for the inconvenience.

[wiak]

Did this occur with beta24 and the Debian kernel 6.0.1?

For me, on beta24, yes (just now upgrading to beta25).
Strange..., cannot explain, fuse is in the list of modprobe'd kernel modules in the init script (in initrd.gz) but at boot it's not loaded.
Simple solution as @geo_c wrote is to put modprobe fuse in a script in ~/Startup or, perhaps make it as a runit service.

Yes, the initrd removes most all unused modules prior to the switch_root to save memory. Then udev loads any new stuff required, but that wouldn't include fuse by default. You have a few ways round this:

1. a startup script
2. a runit service. Don't know details - seen mention of way to do it with systemd here: https://wiki.archlinux.org/title/Udev#M ... s_in_rules
3. force initrd to keep fuse loaded by modifying w_init (the external one will do the job...):
At around line 425, add |fuse:
modprobe -r `lsmod | cut -d' ' -f1 | grep -Ev 'ehci|xhci|sdhci|uas|usbhid'` 2>/dev/null # keep ehci,xhci,sdhci,uas,usbhid
i.e. becomes:
modprobe -r `lsmod | cut -d' ' -f1 | grep -Ev 'ehci|xhci|sdhci|uas|usbhid|fuse'` 2>/dev/null # keep ehci,xhci,sdhci,uas,usbhid,fuse

Note I haven't tried this, but should do what your want.

[rockedge]

I have repackaged beta25 now with Debian kernel 6.0.0-3 SMP PREEMPT DYNAMIC
built with Fred's newest cr-debkernel

Also included wiak's suggestion with the change in line 425 in the external w_init to keep fuse module loaded ->

Code: Select all

modprobe -r `lsmod | cut -d' ' -f1 | grep -Ev 'ehci|xhci|sdhci|uas|usbhid|fuse'` 2>/dev/null # keep ehci,xhci,sdhci,uas,usbhid,fuse

[geo_c]

So is swapping out vmlinuz and intrd all that's necessary, or should I swap all the numbered modules and rootfs? In other words is everything the same as previous except the kernel and initrd?

[rockedge]

@geo_c That's right. To change the kernel only replace the current files to these ->

• 00module-6.0.0-3-amd64.sfs

• initrd.gz

• vmlinuz

• w_init

The external w_init has the fix to keep the module fuse loaded

[fredx181]

...
The external w_init has the fix to keep the module fuse loaded

That works well, fuse problem fixed .
Btw, isn't it best to apply this fix also in the w_init inside initrd.gz ?
(in case someone forgets to include the external w_init )

[rockedge]

@fredx181 I think you're right about putting the fuse mod fix into the initrd.gz

one drawback is that after every kernel build script run the initrd.gz needs to be done manually

[fredx181]

@fredx181 I think you're right about putting the fuse mod fix into the initrd.gz

one drawback is that after every kernel build script run the initrd.gz needs to be done manually

@wiak are you willing to change that in the initrd skeleton ?

[wiak]

@fredx181 I think you're right about putting the fuse mod fix into the initrd.gz

one drawback is that after every kernel build script run the initrd.gz needs to be done manually

@wiak are you willing to change that in the initrd skeleton ?

Yes, Fred, I was planning to do that. Just I'm as slow as the forum at the moment!

[geo_c]

[SOLVED]

So I just used the built-in remaster tool to merge all my changes, now do I replace only the 07KLV-airedale_rootfs.sfs? Or did the 00modules and and 01firmware and 10gtkdialog also get merged? Should I replace all four of those with the new 01KLV-airedale_rootfs-remaster?

Looks like I should because it's numbered 01, which would conflict with 01firmware-6.0.0

I didn't update to @fredx181's latest kernel because jackalpup kept locking up in the middle of the download. I mean totally frozen. Not sure what that was about.

[geo_c]

I used my KLV-airedale20dev install to experiment and figured out that yes, the whole system is merged to 01KLV.

Everything in the remaster appears to be working except a little buggy behavior regarding two symlinks that I have placed in MIME-icons in /root/.config/rox.sourceforge.net which are the inode_directory icons, which I was symlinking for quick changes of the color. So I hardcopied them instead which is just as easy.

Otherwise as far I can tell so far it's all there.

Also my total upper_changes size greatly reduced the entire system size to under 1.8GB, and the system still loads about (edit) 920MB in ram 200MB higher than it did before. So still light and snappy considering everything theme and added application related merged into ram, if I have that correct.

Now I have a base rootfs that's my loaded system. Right now it's running off USB.

I still have this one annoying boot issue with a certain laptop that always labels the usb as sda instead of sdb, and altough all my other puppies and fatdog can deal with that change, it messes with KLV for some reason and won't boot off USB on that laptop, which is a two year old HP-elite i58th gen. No big deal, but it is curious. I wonder why it gets labeled sda1. This happens in fossapup and I believe vanillapup also.

[rockedge]

....the whole system is merged to 01KLV.

Nice! Good experiment.

You can rename 01KLV to 07KLV so not to interfere with 00modules and 01firmware

boot issue with a certain laptop that always labels the usb as sda instead of sdb

We'll look more at why the USB is showing up as sda1. Is there an internal HDD? Is it assigned to a static partition name in /etc/fstab ?

It would be possible to reserve sda1 and assign the USB drive sdb1 by declaring it in /etc/fstab

More info on fstab -> https://www.redhat.com/sysadmin/etc-fstab

[fredx181]

@geo_c Good to hear that the remaster went well, yes only the 01KLV.. sfs (or how you number it) as everything is merged into that.

So still light and snappy considering everything theme and added application related merged into ram, if I have that correct.

Don't know what you mean exactly "merged into ram" it depends on your setup, do you boot with changes in ram w_changes1=RAM2 ?
edit: or copy2ram ?

[wiak]

I my loaded system. Right now it's running off USB.
...
I still have this one annoying boot issue with a certain laptop that always labels the usb as sda instead of sdb, and altough all my other puppies and fatdog can deal with that change

Sounds like you are using w_bootfrom=/mnt/sdX/... method of booting. That won't work if your computer changes the X letter on boot. You should use UUID or LABEL method in your grub.conf or menu.lst

[geo_c]

I my loaded system. Right now it's running off USB.
...
I still have this one annoying boot issue with a certain laptop that always labels the usb as sda instead of sdb, and altough all my other puppies and fatdog can deal with that change

Sounds like you are using w_bootfrom=/mnt/sdX/... method of booting. That won't work if your computer changes the X letter on boot. You should use UUID or LABEL method in your grub.conf or menu.lst

Actually I am using UUID. I could try label instead. But UUID should do it. It's as if that machine is too fast and changes the drive letter or something.

[geo_c]

@geo_c Good to hear that the remaster went well, yes only the 01KLV.. sfs (or how you number it) as everything is merged into that.

So still light and snappy considering everything theme and added application related merged into ram, if I have that correct.

Don't know what you mean exactly "merged into ram" it depends on your setup, do you boot with changes in ram w_changes1=RAM2 ?
edit: or copy2ram ?

Well, I was thinking that the KLV-rootfs was loaded in ram. If that's not the case I stand corrected. I don't currently have any ram options in my stanza. It does seem to be coming in a couple hundred megabytes bigger in ram load. I'll check them later tonight, the unmerged copy against the merged copy.