BookwormPup64 10.0.8

[dancytron]

If I wanted to assist in testing of a new version, how do you remove a test application? Suppose Synaptic brings in 30 individual dependencies when you install a single application in a clean savefile. And then you want to remove that application. How are you gtointatuys doing that?

After removing that (single) application try from terminal: apt autoremove
EDIT: If you use apt (rather than synaptic) to remove that application (e.g. apt purge <package>), there should come some message like: "... ... ... are no longer needed, run 'apt autoremove' to remove these".
(but on BookwormPup, apt may not work as bullet-proof as in pure Debian, not sure though)

While you are at it,

Code: Select all

apt autoclean

is good too. By default Debian likes to save copies of stuff it installs and autoclean deletes that.

[fredx181]

If I wanted to assist in testing of a new version, how do you remove a test application? Suppose Synaptic brings in 30 individual dependencies when you install a single application in a clean savefile. And then you want to remove that application. How are you gtointatuys doing that?

After removing that (single) application try from terminal: apt autoremove
EDIT: If you use apt (rather than synaptic) to remove that application (e.g. apt purge <package>), there should come some message like: "... ... ... are no longer needed, run 'apt autoremove' to remove these".
(but on BookwormPup, apt may not work as bullet-proof as in pure Debian, not sure though)

While you are at it,

Code: Select all

apt autoclean

is good too. By default Debian likes to save copies of stuff it installs and autoclean deletes that.

Note that autoclean (or clean) is for to remove downloaded .deb packages from the cache only.

autoclean: removes all stored archives in your cache for packages that can not be downloaded anymore (thus packages that are no longer in the repo or that have a newer version in the repo).

autoremove: is a whole different thing. it removes software from the system.

EDIT: Btw, AFAIK there's a difference when using apt-get or apt, i.e apt-get install <package> will keep the .deb package(s) in cache and apt install <package> will NOT keep the .deb package(s) in cache.

[Clarity]

Attempted a "qpwgraph" package manager install. Failure shown below:

Bookworm64 installation err.jpg (32.74 KiB) Viewed 2539 times

[fredx181]

Attempted a "qpwgraph" package manager install. Failure shown below:

Installing qpwgraph works for me without error, perhaps something wrong with your setup

[bigpup]

Installed OK for me in BookwormPup64 10.0.4

In synaptic Package Manager

Did you first click the reload icon top left of window?
Let it run and do an update of the latest repository contents.

Then try doing the download and install of qpwgraph.
.

Screenshot(3).jpg (96.02 KiB) Viewed 2491 times

[Clarity]

Thanks guys. I opened a pristine in a VM and agree with your findings.

Thus, something in my production rig causes this problem.

@radky , in the future versions of this, I think the distro would benefit from either Helvum or QPWgraph, OOTB. Pipewire is going to be with us for next decade and this visual ability is going to help everyone.

Again, thanks for all everyone is doing to help this distro.

[OscarTalks]

As with other Debian/Ubuntu pups, if you look in hidden directory /root/.cache you may see that event-sound-cache.tdb* files are created and these accumulate over time. Not the end of the world, but they don't serve any useful purpose. As I understand it, event sounds are not working in Puppy because other packages would be needed for this, but these files arise because Debian/Ubuntu packages of libcanberra are compiled with the --enable-tdb option and libcanberra has to be included to satisfy some dependencies of other packages.

The .pet attached below is the same version of libcanberra but compiled from source with the option --disable-tdb which installs over the top of the Debian version of the libcanberra library and hopefully then prevents the cache files from being created.

Users may wish to test by deleting any event-sound-cache.tdb files that exist, installing the .pet and see if the creation of the files stops. Report any issues and we can respond accordingly. One suggestion for consideration is that senior Woof-CE devs may want to replace the Debian packages with compiled-from-source libcanberra in the recipe so that the Pups don't suffer from this issue out of the box, otherwise users have the option to apply the fix afterwards as described above.

I will attach 32bit versions in the BookwormPup32 discussion thread

[gyrog]

Problem booting a computer because of the kernel in BookwormPup64 10.0.4

Here are my changes to 'DOTconfig' to compile a kernel that works:

Parameters to enablle booting:
CONFIG_MMC_SDHCI_ACPI=y
CONFIG_MMC_SDHCI_PLTFM=y
CONFIG_MMC_WBSD=y

Parameters to enable battery reporting:
CONFIG_AXP288_CHARGER=m
CONFIG_AXP288_FUEL_GAUGE=m
CONFIG_EXTCON_AXP288=m
CONFIG_AXP288_ADC=m

Please include these in the kernel compile for the next version.

[peebee]

Problem booting a computer because of the kernel in BookwormPup64 10.0.4

Here are my changes to 'DOTconfig' to compile a kernel that works:

Parameters to enablle booting:
CONFIG_MMC_SDHCI_ACPI=y
CONFIG_MMC_SDHCI_PLTFM=y
CONFIG_MMC_WBSD=y

Parameters to enable battery reporting:
CONFIG_AXP288_CHARGER=m
CONFIG_AXP288_FUEL_GAUGE=m
CONFIG_EXTCON_AXP288=m
CONFIG_AXP288_ADC=m

Please include these in the kernel compile for the next version.

The BW64 kernel is built on Github Woof-CE by applying "diffs" to the Debian kernel....
https://github.com/puppylinux-woof-CE/w ... s/bookworm
would need to be updated (I think).

[dimkr]

Problem booting a computer because of the kernel in BookwormPup64 10.0.4

Here are my changes to 'DOTconfig' to compile a kernel that works:

Parameters to enablle booting:
CONFIG_MMC_SDHCI_ACPI=y
CONFIG_MMC_SDHCI_PLTFM=y
CONFIG_MMC_WBSD=y

Parameters to enable battery reporting:
CONFIG_AXP288_CHARGER=m
CONFIG_AXP288_FUEL_GAUGE=m
CONFIG_EXTCON_AXP288=m
CONFIG_AXP288_ADC=m

Please include these in the kernel compile for the next version.

I'll open a PR.

EDIT: https://github.com/puppylinux-woof-CE/woof-CE/pull/4229, trigger https://github.com/puppylinux-woof-CE/w ... kernel.yml after this is merged, give it 2-3 hours and you'll get a kernel you can test.

[peebee]

https://github.com/puppylinux-woof-CE/w ... 05cd4ffa85

[gyrog]

https://github.com/puppylinux-woof-CE/w ... 05cd4ffa85

I appreciate that you have reacted to my suggestion,
BUT sorry, this doesn't look like it's worth it.
Why have you not included the config lines to enable the kernel to see the battery?

This computer is a small laptop, as such, you need to be able to see that status of the battery.
If the battery configs are not there, I'm still going to have to compile the kernel, to properly support this computer.

[gyrog]

@peebee, @dimkr,
Sorry, it looks like I was a bit hasty, in my appraisal of the PR.
Looking at the DOTconfig in the zdrv in the Bookworm64 ISO, it looks like the AXP288 configs are already present.

So looking in woof-ce I find that "dpup-kernel" was run by "peabee" 2 days ago,
is this the updated kernel?
Or should I wait for another?

[dimkr]

You'll need to trigger a build or wait until the next one.

[gyrog]

@dimkr, and @peebee, Thankyou.
The new kernel works in my Lenovo.

[retiredt00]

I was wondering if you happened to see these posts regarding laptop suspend on lid closure
https://forum.puppylinux.com/viewtopic. ... 32#p107932
https://forum.puppylinux.com/viewtopic. ... 60#p108560
https://forum.puppylinux.com/viewtopic. ... 69#p108569

Dear radky and dimkr
the reason that bookworm and vanilla dpup fail to suspend on lid closure is that they miss the xserver-xorg-input-evdev package.
Adding it through the package manager, solves the problem
Please consider adding it in future builds

[vtpup]

If I wanted to assist in testing of a new version, how do you remove a test application? Suppose Synaptic brings in 30 individual dependencies when you install a single application in a clean savefile. And then you want to remove that application. How are you guys doing that?

After removing that (single) application try from terminal: apt autoremove
EDIT: If you use apt (rather than synaptic) to remove that application (e.g. apt purge <package>), there should come some message like: "... ... ... are no longer needed, run 'apt autoremove' to remove these".
(but on BookwormPup, apt may not work as bullet-proof as in pure Debian, not sure though)

Thanks Fred and Dancytron. There are some important differences between the way the Puppy Package Manager handles program removals and how Synaptic does, and apt and apt-get do.

1.) PPM quickly and easily handles a program removal with dependencies in a graphical user interface. For dependencies that it is unsure of, an informational window is displayed showing those dependencies. It does not remove those. This is a smart program manager.

2.) Synaptic is the graphical program installer for BookwormPup. It does not remove any dependencies. This is a major drawback. It has two named functions "removal" and "complete removal" Neither actually does more than remove a program, leaving the dependencies "Complete only removes configuration files. It is a misnomer. This is not a smart program manager.

4.) Apt and apt-get are console commands, not graphical program managers. The apt commands provided for program removal are not intuitive, What does remove, autoremove, purge, clean, autoclean mean to a normal user without looking at a man page for each?

5.) None of the apt and apt get commands do what the PPM does.

My understanding: Apt autoremove attempts to remove ALL orphan dependencies in a system, not just the unused dependencies for a specific installed program. It is unclear how it establishes which dependencies are orphaned and which are needed. It does not seem to track what was downloaded as dependencies for a specific program. It is not program oriented, it is system oriented. It provides no information about possible conflict dependencies for a particular program as Puppy does. Maybe it just keeps possible conflicts, maybe it doesn't, who knows? It also does not remove downloaded programs held in a program cache. The PPM does remove those.

To remove cached downloaded program files you also have to run apt purge but this also removes configuration files, which you may not want to do. In many cases it is desirable to retain a config file, even though a program is removed. For instance when reinstalling for some reason after personalizing settings..

Apt clean clears out the entire local repository of retrieved package files. The PPM does not do this it only removes the specific retrieved program file

Apt autoclean only removes package files that can no longer be downloaded from a repository. PPM has no equivalent, and in fact you may not want this if there's a program you want to keep a local copy of because it has disappeared from an online repository. You don't necessarily want a program removed just because it is removed from an online repository, or replaced with a later version.

Clean and autoclean are actually local repository maintenance commands. They either remove all local copies of downloaded programs, or remove those no longer mirrored.

For me, these commands are problematic for normal users, and may cause personal system problems. It seems to me more likely that autoremove, for instance will accidentally remove an essential dependency somewhere else in a system than PPM, because it is a wholesale dependency removal tool, not a specific program dependency removal tool. I see lots of users on apt oriented systems having problems after cleaning their systems, more than than puppy users removing a single program with PPM.

I really think we're giving up a more mature package management system in PPM, for Synaptic's lack of program management capability, and the requirement for an additional collection of confusingly named and oddly acting console commands. I think a user should be able click a "Remove Program" button in a graphical package management app to restore an efficient system state before the installation of a program with multiple dependencies. And I would guess most actually think that happens in Synaptic. It just doesn't..

[rockedge]

I really think we're giving up a mature package management system in PPM

Pkg does a far better job than PPM overall and is far more powerful. In PPM it is close to impossible for an average user to add or change repositories. PPM does a poor job in dependency checking on any complicated packages.

[ozsouth]

Unless ppm packages are carefully crafted, removal is a real problem. Not many use puninstall.sh for clean removal. With my chromebook under crostini (debian bullseye), 'sudo apt remove && sudo apt autoremove' is very effective.

[vtpup]

Well that's true Rockedge re changing reposirories. I'm not a PPM lover -- but it sure does a better job of managing what it does download than Synaptic.

Not sure what "Pkg" is...?

[vtpup]

Okay ditch PPM, but do we then really want new users to revert to console commands? If so better put up informational modifications to Synaptic to let users know its "remove" and "complete remove" functions don't really do what they say, and then explain how to enter the console apt commands, and what each does, and the syntax required.

Feels kludgy to me -- and I'd bet new Puppy tryouts as well.

[bigpup]

BookwormPup64 has this:

Main menu ->Help ->HOWTO use the APT Package Manager

This provides a how to explaining what command does what and how to use it.

[vtpup]

BookwormPup64 has this:

Main menu ->Help ->HOWTO use the APT Package Manager

This provides a how to explaining what command does what and how to use it.

A.) Puppy Linux is a GUI OS. Like most other OS's in the last forty years.
B.) Synaptic is Bookworm's GUI package management system, Synaptic, has improperly named and ineffective package removal functions.
C.) A new user should not have to guess that some function in Synaptic doesn't do what it says it does, then figure out that they should look for a help file to find out how to do via console commands what Synaptic said it already did.

The actual result of B.) will be that new users will get bloated systems and systems with problematic dependencies that they don't know exist, and which interfere with other programs and functions. And that kind of issue will appear frequently here on the forum. Or will silently turn off potential new users to Puppy as an unreliable OS.

[vtpup]

I've been experimenting with package management systems this morning -- adding and removing programs by a variety of methods and looking at what files are removed and what aren't.

1.) I added a program with Synaptic. 50 packages were downloaded and installed.
2.) In Synaptic I marked it for "complete removal", then executed. Result? 1 package was removed
3.) I opened up console and entered apt autoremove. Result? 14 packages were removed.
4.) 35 packages were not removed.

Other interesting experiments:

I installed an alternative deb package management system called : gnome-software

This installation took 75 megabytes, so it was large by Puppy standards, however, it was far superior as a user friendly GUI to Synaptic. It provided much more usefully worded app information than Synaptic does. The interface was well organized, attractive, and easy to use. It also did a much better job of removing an application that it had installed earlier.

While like all of the above methods, it wasn't perfect at removal, it was far more effective than Synaptic's single file removal, and actually appeared to be on par with apt autoremove.

Interesting point: while it is 75 megs large, if it removed a program more effectively than Synaptic, it would very quickly make up for its size by reducing bloat. Perhaps even with the first program it removes.

[wizard]

BookwormPup64 10.0.4

Been doing some test on bootable USB creation and found some methods/conditions that will fail.

Condition 1 - HDD/SSD and USB have a partition with the same name
-USB partitoned with Gparted
-install Puppy using Menu>Setup>Frugalpup>Puppy (this install creates a "Savespec" file)
-install grub2 using grub2config 2.01
Result:
-When the USB, that has a save folder, is booted it a cannot find its savefolder.
Resolution:
-delete or rename the Savespec file or rename the partition on the USB

Condition 2 - USB install using Menu>Setup>Frugalpup>Puppy, then grub installed with Menu>Setup>Frugalpup>Boot
-USB partitoned with Gparted
-install Puppy using Menu>Setup>Frugalpup>Puppy
-grub install with Menu>Setup>Frugalpup>Boot
Result:
-USB fails to boot
Resolution:
-none, use different method

Thought I'd post these since it might help when troubleshooting USB boot issues.

wizard

[dancytron]

Hardware report.

I bought a $32 ebay Gen 3 Thinkpad 11e with a Ultima motherboard.

https://psref.lenovo.com/syspool/Sys/PD ... n_Spec.PDF

I set it up for dual boot (I'll post how to do that later) edit: see viewtopic.php?p=110957#p110957 and set up a usb boot disk with old style bios boot using the Debian Dog Starter Kit method.

It boots, recognizes all the hardware and everything works with one wrinkle. The default touchpad settings make it appear that the mouse is frozen and just teases you with moving a little. Plug in a regular mouse, go into the settings app, and turn some features off and make it more sensitive and it works.

Works fine and is fast enough for both Puppy and MX with the stock bookworm kernel for $32.

Dan

edit: also microphone doesn't work

[OscarTalks]

BookwormPup64-10.0.4
I see that multimedia playlist files with the suffix .m3u and also .m3u8 show up in ROX as plain text files when really they should usually be set to be opened by a media player (although you should be able to right-click and open as text if you want to).

There may be a better way of rectifying this, but I have found that the file:-
/usr/share/mime/packages/freedesktop.org.xml
has between lines 4614 and 4625 inclusive the following:-

Code: Select all

  <mime-type type="application/vnd.apple.mpegurl">
    <comment>Media playlist</comment>
    <sub-class-of type="text/plain"/>
    <glob pattern="*.m3u"/>
    <glob pattern="*.m3u8"/>
    <magic priority="70">
      <match type="string" value="#EXTM3U" offset="0">
        <match type="string" value="#EXT-X-TARGETDURATION" offset="0:128"/>
        <match type="string" value="#EXT-X-STREAM-INF" offset="0:128"/>
      </match>
    </magic>
  </mime-type>

This seems to be hijacking the mime-type (audio/x-mpegurl) and preventing it from displaying correctly in ROX
So I deleted those lines of code from the file and saved the modification
Then in terminal I ran:-
update-mime-database /usr/share/mime
Those playlist files now show up as media files with correct right-click options and are opened with default audio player on left-click

[mikewalsh]

I installed an alternative deb package management system called : gnome-software

This installation took 75 megabytes, so it was large by Puppy standards, however, it was far superior as a user friendly GUI to Synaptic. It provided much more usefully worded app information than Synaptic does. The interface was well organized, attractive, and easy to use. It also did a much better job of removing an application that it had installed earlier.

While like all of the above methods, it wasn't perfect at removal, it was far more effective than Synaptic's single file removal, and actually appeared to be on par with apt autoremove.

Interesting point: while it is 75 megs large, if it removed a program more effectively than Synaptic, it would very quickly make up for its size by reducing bloat. Perhaps even with the first program it removes.

Heh. I've been saying for years that Synaptic is NOT all it's cracked up to be. I migrated to Puppy to get away from it.......and now it's been chosen as the PPM replacement? Puh-leeze.....

Synaptic was always the "neckbeard's" choice.....but it's neither intuitive NOR "easy-to-use". To my way of thinking, one of the main reasons it became so universally employed at one point was due to the way many distro 're-spins' were put together.

A lot were Ubuntu-based. Canonical had decided Synaptic was the way to go at one time, so every 're-spin' out there had it too. I sometimes got the impression that if Canonical had decided to heave their OS over a cliff edge, then the 're-spins' would have followed that as well....

I agree we do need a decent GUI package manager.....but there's no denying that command-line installation is faster and more versatile. MyCrudSoft's over-riding legacy will always be remembered as teaching the world's population that if they DON'T have a GUI for something, then it can't be used......(??!)

My view is this; I think that so long as we have a fully-functional apt installer mechanism, along with an easy way to add/remove extra repos, Puppy will be flexible enough to survive the future.

Just my two-penn'orth, FWIW.

Mike.

[dimkr]

and now it's been chosen as the PPM replacement? Puh-leeze.....

PPM is unmaintained for years, it's unreliable (especially when used with Debian packages) and it makes it very easy to break your system.

Anyone who wants PPM back should either:

1. Volunteer to maintain PPM: fix known bugs, fix its architecture (for example, make it understand a situation where package a depends on b or c - if user installs a but c is present, no need to install b), make it less CPU intensive and make it faster, or
2. Find somebody else to do this

[bigpup]

WELCOME TO LINUX SOFTWARE!

I never understood why people wanted to drop using PPM as the Puppy package manager.

It does what it does very well.

However, it is limited and really does need to have some tweaking and features added.
Being able to easily add a repository, to get software from, is one.

PPM is unmaintained for years, it's unreliable (especially when used with Debian packages) and it makes it very easy to break your system.

This is mainly because the deb package is coming from a repository, that is for a specific other than Puppy Linux OS (say a version of Ubuntu).
The deb package works OK for the Ubuntu version, because it is compiled for that specific version of Ubuntu, and what it has already in it.
It is a crap shoot, it may also work in the Puppy version, based on that specific version of Ubuntu.
Some do work, some do not.

With a little work, by someone that understands the program code.
PPM could be still a very good package manager.
Maybe have a way it could actually use apt and record exactly what apt did.
So, uninstall in PPM could do a good job, to uninstall something, apt installed.

APT:
Having apt added to Puppy, as a way to get software, has been long needed, but it too is limited, on how well it does work.
But if you go find software on the Internet, all that it provides, is apt commands to get the download of it, and how to install.
Apt is needed to be there to use.
I have gotten software from program web sites by using apt.
The program was not offered any other way, but using apt commands.

Using apt to get software, that is compiled for a specific version of say Debian and is kernel specific stuff, does not work in Puppy, unless it too is using the exact same kernel.
So having apt access Debian repositories for software, does not work for kernel specific software, like any kind of drivers. (Graphic, WIFI, hardware, etc....)

TO LINUX SOFTWARE!

Even pet packages, offered on this forum, do not work in all Puppy versions!