Lucid Puppy 5.2.8 - Updated ISO Version 005 - APR 05 2012

[bigpup]

Today I removed two partitions from one drive of my Puppy system. They were named sda1 and sda5, originally from a Windows partitioning system.

I think your problems started because you removed the first partition on the hard drive sda1.
That probably screwed up the partition table and so, explains why partition programs are giving strange output.
Gparted, in Puppy, is a little conservative on what it will do.
Parted Magic should be more aggressive.

Normally, partitions follow a logical progression of, start at the beginning, of a hard drive, and go towards the end. You deleted the beginning partition and threw everything off.

Resizing and deleting of partitions, always works best, if you start at the end and work towards the beginning of the drive.

[Brown Mouse]

Pmusic 2.4.4-1
http://www.murga-linux.com/puppy/viewtopic.php?t=31206

I've updated Pmusic from 2.30 to 2.4.4-1 but now when I launch a music cd I don't get any track information(apart form the no.) or album artwork as before.
Anyone else see this?

Yes, I am seeing this too.
This may not be a good update, but only way to know is try it.

I am seeing other problems.
Like change the view, and there is no way to change back, that I can find, on some of the optional views.
I am going to post about tT
his on the Pmusic topic.

Thanks for the confirmation.I have reverted to 2.30 for the time being.

[ICPUG]

In http://bkhome.org/blog/?viewDetailed=02740 Barry says the following:

The best way to update a puplet is to build from latest Woof. Wary, Racy, Slacko, and some others, are doing this. A version upgrade doesn't change what it is, a Slacko based on Slackware 13.37 remains as that.

Lucid is in a different category as it forked from Woof sometime ago. Lucid upgrades do not sync with the official Woof. Instead, Lucid maintainers may, if they wish, pull out whatever improvements in Woof that they want.

Could Playdayz explain in simple words a non-Woof user can understand how Lucid has forked from woof.

I can understand how it might be based on an old version of woof and therefore not include the latest bugfixes. What I do not understand is how it is forked from woof and what the implications are in doing so.

[otropogo]

[

quote="bigpup"]

Today I removed two partitions from one drive of my Puppy system. They were named sda1 and sda5, originally from a Windows partitioning system.

I think your problems started because you removed the first partition on the hard drive sda1.
That probably screwed up the partition table and so, explains why partition programs are giving strange output.

I wish it were so. But Gparted has failed to open on lupu 5.2.8 as long as I can remember, certainly months before I removed any partitions.

And I only went to PartedMagic after finding I was unable to access the partitions with both of the lupu partition utilities.

Resizing and deleting of partitions, always works best, if you start at the end and work towards the beginning of the drive.

Thanks for the tip. But that wasn't possible in this case, as the one partition I needed to preserve was at the end of the drive.

[playdayz]

Laptop Tools by pemasu

If you have a laptop and want the fancy stuff, this looks good
http://www.murga-linux.com/puppy/viewto ... 125#563125

--------------------------------------------------------------------------

Could Playdayz explain in simple words a non-Woof user can understand how Lucid has forked from woof.

I can understand how it might be based on an old version of woof and therefore not include the latest bugfixes. What I do not understand is how it is forked from woof and what the implications are in doing so.

Lucid 5.2.8 was built with a particular version of Woof--the "last non-zzz" version to be precise. After that Barry made significant changes to some of the basic infrastructure of Woof. I think you state the situation accurately "based on an old version of woof and therefore not include the latest bugfixes." However, some important bugfixes from later versions of Woof have been incorporated into Lucid and there have been many Lucid-specific bugfixes as well.

OK, so why did I stop? Because it was done! I mean, it had to stop somewhere, and the last non-zzz Puppy seemed like a good milestone. Every rebuild with Woof necessitated a significant round of testing and after a year and a half of 4-6 hour days, I had some other things to do. I believe Barry would be happy to see someone start with the latest Woof and the latest Ubuntu and make a brand new Upup.

[radky]

Updated Programs:

PupApps-1.8
http://murga-linux.com/puppy/viewtopic.php?t=61962

PupControl-2.0
http://murga-linux.com/puppy/viewtopic.php?t=61961

PupShutdown-1.8
http://murga-linux.com/puppy/viewtopic.php?t=56077

[radky]

New Program:

PupMenu-1.0
http://murga-linux.com/puppy/viewtopic.php?t=76713

[rerwin]

playdayz and testers,
I am attaching cumulative patch-2 for the testing version of lucid pup. It contains everything in patch-1, described here:
http://murga-linux.com/puppy/viewtopic. ... 374#607374
plus the following:

• - Comprehensive integration of support for Broadcom bcm43xx wifi devices, including the proprietary (wl) driver, to prevent conflicts and avoid the issue of b43 loading too early. No blacklisting is needed unless you need to force use of the open-source driver (by blacklisting wl).
  - From woof: Universal Installer fixed - Posted on 10 Mar 2012: npierce: some grep operations were wrong in puppyinstaller, if have more than 9 partitions
  - From woof: BootFlash fix - Posted on 11 Mar 2012: BootFlash: 128m 1st partition increased to 256m, to handle bigger puppies
  - From wary 5.2.2: SNS version of rc.network that contains additional logging

Note that my fixes ported from wary and woof exclude the recent updates related to localization, as that support is still in development and might be risky to add to lucid pup until further research is done.

Regarding the Broadcom wifi fix: It is made possible by peebee's testing support -- seemingly endless cycles of install something, run a bunch of reboots, gather diagnostic information, and send them to me. After initially reporting that on some reboots, the previous connection was not being re-established, he worked with me as I tried various experiments. I could not have done this without his help, as he has the hardware that uses the proprietary driver. My thanks go to him.
Richard

[Volhout]

Rerwin,

some 6 weeks back I tested the 528TESTING01 iso, and reported a problem with my USB printer/scanner, not detected by CUPS.
Later PlayDayz withdrew the testing iso, and told everyone to continue with 004.

In all previous 52x's (528 tested up to -004) CUPS worked flawless with my printer.

I am not sure if you looked at it, and if you found a fix.

Regards,

Volhout

[ICPUG]

Thank you Playdayz for taking the time to respond to my query.

Everything has to stop somewhere because there are continual developments of everything - woof - applications - desktops(!) - and sometimes these developments go veering off into unknown territory (gnome 3 and unity spring to mind but lucid had to cope with kernel mode setting and the autoconfiguring X-Org!)

The Puppy development model doesn't lend itself to trying to keep up to date with everything like the major distros do. On the other hand, the Puppy development model creates some really great useable distros for all types of user. As someone of the 'if it ain't broke don't fix it' brigade I am quite happy to use Lucid until it doesn't work for me!

[LateAdopter]

Hello otropogo

Since this is way off topic I will be very brief.

The Win2k installer does not understand 48 bit addressing. You get that error message when trying to install with any partition involved which goes past 137GB ( including the extended container). You also have to EnableBigLBA in the registry after install for 48 bit addressing to work.

The entries in the partition table can be in any order. gparted does them in the order you entered them regardless of where the primary partitions are on the disk. But windows disk manager will reorder them if you use it to modify anything and the sda numbers will change.

[otropogo]

Hello otropogo

Since this is way off topic I will be very brief.

The Win2k installer does not understand 48 bit addressing. You get that error message when trying to install with any partition involved which goes past 137GB ( including the extended container). You also have to EnableBigLBA in the registry after install for 48 bit addressing to work.

The entries in the partition table can be in any order. gparted does them in the order you entered them regardless of where the primary partitions are on the disk. But windows disk manager will reorder them if you use it to modify anything and the sda numbers will change.

Hi LateAdopter,

Thanks for your explanation. But I'm not quite clear about your statement:

You get that error message when trying to install with any partition involved which goes past 137GB ( including the extended container).

because:

1. initially I tried to install only to a primary partition modestly sized at 10GB, NOT >137GB

2. eventually I tried to install to the second hard drive, which contained no partition as large as 137GB,

Did you mean to say that Win2K can't install to any hdd larger than 137GB? Or perhaps to any hdd with a contiguous unallocated area >137GB?

And I feel it IS a lupu5.2.8 issue, in that all of the available partitioning tools failed to provide any support.

The fact that I had to resort to a PartedMagic installation booted from a rescue disk is, I feel, a fact demanding very serious consideration. The fact that PartedMagic reported no issues with either drive or any of their partitions, makes the failure of all of lupu's bundled partition managers all the more significant.

The other issue, which may be outside the scope of this thread, as you suggest, is the fact that Win2K failed to recognize the ntfs partition freshly created by PartedMagic.

But you haven't addressed that issue...

[LateAdopter]

Hello otropogo

I have installed win2k and various puppies to a 320GB SATA drive. I partitioned it using the gparted that is in Lupu 525.

I don't know where the win2k installer goes wrong exactly, but it will not understand any sector address on the disk that is more than 137GB. So I speculate that anything that involves an address higher than that will fail or produce unpredictable results.

In my case I was installing to a 5GB extended logical partition that was all below 137GB on the disk, but it failed. The extended container partition defined in the MBR went right to the end of the disk i.e.320GB. I reduced this to below 137GB and then the install worked. Having done EnableBigLBA in win2k I then used gparted to increase it back to its full size again.

It was when I used win2k disk manager to create and format two ntfs extended logical partitions that it reordered the entries in the partition table in the MBR. Which was confusing to puppies and humans alike!

[bigpup]

The other issue, which may be outside the scope of this thread, as you suggest, is the fact that Win2K failed to recognize the ntfs partition freshly created by PartedMagic.

This is normal with Windows. Especially Windows installer program. The only ntfs partitions that it will recognize are ones made by Windows partition programs. Third party programs, you pay for, get the secret code Windows is looking for on the partition.
I have read reference to some kind of security code Windows wants to see.
Gparted will handle ntfs format, but being open source, it does not have 100% support from Microsoft, like pay for partition programs do. It does not give Microsoft money for the secret code Windows looks for.
Any time you use Gparted to re-size a ntfs partition that Windows is installed on, you have to use Windows boot disc or repair disc to do a repair operation.
Explained here:
Talks about Vista and Windows 7, but holds true for any Windows.
http://www.howtogeek.com/howto/windows- ... partition/

[otropogo]

Thanks bigpup,

This may explain many, perhaps all, of the major Windows problems I've had over several years, not just with Gparted, but perhaps also from using Ranish Partition Manager.

[otropogo]

Hello otropogo

I have installed win2k and various puppies to a 320GB SATA drive. I partitioned it using the gparted that is in Lupu 525.

I don't know where the win2k installer goes wrong exactly, but it will not understand any sector address on the disk that is more than 137GB. So I speculate that anything that involves an address higher than that will fail or produce unpredictable results.

In my case I was installing to a 5GB extended logical partition that was all below 137GB on the disk, but it failed. The extended container partition defined in the MBR went right to the end of the disk i.e.320GB. I reduced this to below 137GB and then the install worked. Having done EnableBigLBA in win2k I then used gparted to increase it back to its full size again.

It was when I used win2k disk manager to create and format two ntfs extended logical partitions that it reordered the entries in the partition table in the MBR. Which was confusing to puppies and humans alike!

Encouraging news, but unfortunately it makes the error messages and lockups I got during installation even more mysterious. From your experience it follows that the size of any unallocated space is not a factor. So I should have been able to install to my pre-created ntfs partition.

Of course I didn't use lupu's Gparted, since it won't open (ever) on my system. Nor could I use pdisk's utilities, as fdisk warned of a possible problem (on both hdds) with the cylinder count being thirty times the "normal" maximum 30401 vs 1024),

I wonder if anyone else is getting this error message from pdisk/fdisk on drives larger than 139GB?

and cfdisk warning for hdd0,

FATAL ERROR: Bad primary partition 2:partition ends in the final partial cylind
Press any key to exit cfdsk

The line cannot be scrolled, to read the portion that's cut off, nor can the window be maximized or resized. Any of these attempts shuts down pdisk.

With hdd1 cfdisk performs normally. But these inadequacies make me leery of using either utility. It seems to me that to be usable, a disk utility should at least be able to meaningfully identify any problems with a drive and preferably be able to fix them or at least suggest another solution. The fact that the PartedMagic rescue LiveCD reported no issues with either drive's partitioning system or cyclinder count does nothing to allay this concern.

[kevin bowers]

Of course I didn't use lupu's Gparted, since it won't open (ever) on my system. Nor could I use pdisk's utilities, as fdisk warned of a possible problem (on both hdds) with the cylinder count being thirty times the "normal" maximum 30401 vs 1024),

I wonder if anyone else is getting this error message from pdisk/fdisk on drives larger than 139GB?

There was a problem in DOS with drives >32GB, and that is where that error warning came from. I thought it was fixed from WinME on, but perhaps Win2K missed it? Microsoft issued a hotfix for Win98, but it caused more problems than it solved, as the actual problem was just a technicality and didn't affect the usability of the drive or system. All it amounted to was that FDISK reported some sizes wrong, and it could safely be ignored, so I wouldn't worry about that.

[otropogo]

...

There was a problem in DOS with drives >32GB, and that is where that error warning came from. ...All it amounted to was that FDISK reported some sizes wrong, and it could safely be ignored, so I wouldn't worry about that.

What worries me is that pdisk/fdisk reports one problem, while pdisk/cfdisk another. Evidently, one of these modules is defective, perhaps both.

Not something I want to bet my data on.

The fact that lupu's Gparted won't run at all, worries me too. It suggests that there's something about my hard drive configuration that none of them are equipped to handle.

[ICPUG]

IF
you want to create/modify/delete ntfs partitions that are to be useable
by windows and linux
THEN
use Windows partitioning tools
ELSEIF
you want to create/modify/delete ntfs partitions that are to be useable
by only linux
THEN
you can use Linux partitioning tools
ENDIF

Once you stray from this logic then behaviour becomes unknown and perhaps it is better to start from scratch - partition with the right tool and reinstall your operating systems.

My ntfs partitions are always useable by both Windows and Linux so I use Windows partitioning tools.

Once the partitions are there I have found writing to them with Linux to be no problem despite the dire warnings of the Knoppix community (and the occasional commentator here)!

This is a problem not specific to Lupu. It should be in a more general forum.

[Sylvander]

A.

IF
you want to create/modify/delete ntfs partitions that are to be useable
by windows and linux
THEN
use Windows partitioning tools
ELSEIF
you want to create/modify/delete ntfs partitions that are to be useable
by only linux
THEN
you can use Linux partitioning tools
ENDIF

I discovered an example of the truth of this quite some time back, as follows:
1. I used Puppy->Gparted to create a FAT32 partition, on which to install Win2000Pro...
BUT...

2. I discovered during the installation of Win2000Pro...
At the point where all the files SEEM to have been installed...
And there is a reboot to load the Windows installation...
The Windows SETUP goes right back to the start, and begins afresh.
This happens in a continuous loop.
Win2000Pro would NEVER have finished the installation process.

3. It seemed the only way I could fix it, was to partition and format using Microsoft programs.
I used a bootable floppy->[Bob Cerelli's Win98 Startup Floppy], and ran fdisk.exe and format.com

B. Another time I discovered:
When I used Puppy->Gparted to non-destructively reduce the size of an NTFS partition holding an installation of WinXP...
XP would no longer boot!
I had to use TestDisk to recover/repair the Windows partition, so XP would boot once again.