Some Puppy ISO's won't boot

[CatDude]

Hi

I thought you had other Puppies that do boot OK

Bootup using one of those, then mount the partition where the ISO images reside,
navigate to the directory containing the ISO images and check the md5sums.

CatDude
.

[bert07]

Somebody else also asked for giving in codes.

Does not work here. Stops loading (?) at boot time.

[bert07]

Only on the compaq.

[disciple]

I do not think it is kernel related (as such), because versions run on my other computers.

FWIW that's the reason why people think it is likely to be kernel-related.

Someone was blaming it on a fault with the multisession system. But are you actually using a multisession Puppy CD?

[ocpaul20]

So, it seems that it is a compatibility problem between the hardware on the compaq and the software for Lupu and quirky.

Something which may cause problems which might be related to this problem is that DVDs read on one machine sometimes do not work very well (read errors or not load correctly) if they have been burned by another machines DVD drive. I believe there may be some tracking issue which causes this to occur.

It may be a good idea to burn the DVD on the machine that you intend to read it from, so that the write and read are done using the same DVD drive. Also some cheaper DVDs do not burn so well too. I have found this in the past.

Just something else to consider.

[Béèm]

Although Quirck 1.4.0 and as far as I can remember Lupu 5.0.0 run fine on that machine, Quirky 1.42 and Lupu 5.2.5 and 5.2.5 do not.

This is what bert did write in his first post (I corrected the typos)
So Quirky 1.4.0 do work on that Compaq, Quirky 1.4.2 doesn't
Lupu 5.2.0 do work an that Compaq, Lupu 5.2.5 doesn't.

However Quirky 1.4.2 and Lupu 5.2.5 work correctly on bert's 4 other PC's.

In the long discussion we had, bert told also that he has the kernel panic already before he gets the splash screen, so he can't enter a boot code puppy pfix=XXXXXX. In all the years I follow here, I never saw this issue mentioned.

I wasn't able to determine if Quirky 1.4.2 and Lupu 5.2.5 have the same kernel and if that one is different from the Quirky 1.4.0 or Lupu 5.2.0
If different, that could be an indication.

I also advised bert to PM Barry as this issue is too weird.
I am not sure he did so already.

Hope this brings the everybody at the same level of understanding.

[disciple]

I don't run Puppy from CD, I just do a frugal install straight from the .iso

Can anyone confirm whether the kernel is actually loaded before the boot menu when running from CD? It certainly isn't loaded before the boot menu when booting from hard disk using grub, and I thought it would be the same booting from CD... if so, then it would be impossible to get a kernel panic before the boot menu.

[Béèm]

I run the same way as disicple. So no more CD boot regularly.
To my knowledge vmlinuz is the kernel to load.
Loading can be influenced by:
1 - if booting from grub, with the parameters given on the kernel line in it
2 - if booting from CD by typing in the 5 second pause puppy xxxxxxxx

So I find it very strange also that bert claims he has a kernel panic before that 5 second pause. Following is the contents of isolinux.cfg

Code: Select all

default puppy
display boot.msg
prompt 1
timeout 50

F1 boot.msg
F2 help.msg
F3 help2.msg

label puppy
kernel vmlinuz
append initrd=initrd.gz pmedia=cd

boot.msg contains

Code: Select all

logo.16
0aJust wait 5 seconds for normal startup!07
For basic help press the <F2> key, for advanced help press <F3> key.

So it is clear That the 5 seconds prompt is there before the loading of the kernel.

[BarryK]

bert07's posts are too confusing for me. As Béèm has pointed out, you can't get a kernel panic before the splash screen. Also, bert07 was asked to do a md5sum check, which he does not seem to have done.

[bert07]

Checksums

b6658ab75cd5d48f358f0ee31b06b934 lupu-525.iso

2cfd4279b20a008feee0dde205d6b8c1 qrky-142-all-drvrs.iso

---+---
I must say I almost have it working but ran into another problem:

I did a CD rebuild (with the build in tool) save the ISO, burned it with Pburn, and now Puppy 5.2.5 does boot and install on the compaq (also loads lots faster on the other ones).
But now I have a screen problem: the screen does look very "vague" as if not cleaned for 5 years. And that's on all my computers.

Sigh.

[Béèm]

Run xorgwizard again and select x16, not x24

[bert07]

I did run 16bit, but I solved the screen problem: I just need to use VESA now.

So all in all: the entire problem (for me) is solved by rebuilding the cd and than using VESA as the graphic driver.

I can boot and install Puppy 5.2.5 now.
(Boots way faster than the original ISO.)

Thanks to all out there!


And again sorry for my (at a time) not so correct behaviour.


Thanks again!

[gcmartin]

I have provided a significant amount of information on the problem of "Save-session disables the media from subsequent booting."

Questions for forum members

• If I make an ISO of the media, is there anyone in this forum who understands boot media well enough to review it for problems?
• If so, where should I post the ISO?
• Would it be acceptable to use "dd if=/mnt/sr0 of=wontboot.iso" be an acceptable method of presenting this for forum review?
• Should all the information provided in the other post be "zipped" into a package with the ISO so that its all in one place?
• Would anyone indicate whether this is acceptable?

Thanks in advance
P.S. Can anyone explain what makes a data media bootable for Linux OSs? Does the file location on CD have something to do with this or is it some other technology so that the CD/DVD device detects bootable media? (This might help in understanding how save-session is destroying the bootable live media.)

[Béèm]

bert07.
Indeed using the vesa driver is a solution. I forgot about that one.
Success and don't hesitate to post.

For the sake of fellow poster can you edit the subject of your initial post and add [SOLVED] at the end.

[bert07]

@gcmartin

I'm way to "stupid" to run most Linux commands, meaning I run Linux in a graphic environment and I can't remember most "terminal" codes.

I do not understand the way an ISO build works, but considering that the "build-in" remastering works, surely tells us that there is something wrong with the original build.

I also need to try this for the quirky 1.4.2 build, but (before I tried this) I can almost be sure that in that build "my" problem will be solved doing just the same and remaster the CD (ISO).

I still need to try it. Will do so tomorrow or the day after.

What the latest ISO's are concerned, I feel that the "coders" need to have a look at it. There seems to be something wrong with it one way or the other.

No offence.

[bert07]

@ Béèm
I could end this thread with [Solved], but that is not really true.

It is solved for me, or I ran "around" the issue, but the "issue" still exists. It is not solved.

Unless you mean "solved for me".

---+---
Also I noticed that many a distro build (burned on CD or DVD) is not "recognized" as easily as it is supposed to.

Might be an issue that perhaps is not relevant here, but with an al-around issue for CD/DVD booting.

---+---
So I won't put "solved" at the end, but perhaps this issue should go to another thread.
Also because I noticed afterwards that the remastered ISO boots much faster on my other computers as well.

Something is not right here.

If it should go to another thread, pray let me know here, so I can follow up the issue.

[nooby]

GCMartin that is very interesting. Rcrsn51 made a pet that allowed me to boot Peppermint OS on a USB flashmemory and it worked very good and then I tried to boot isos and puppies on the NTFS and then the booting of the iso on the USB stopped to work. Could it be same phenomena? The Peppermint could have saved something to the USB that made it refuse to boot?

I start all over tomorrow to see if it boot once again and then twice or if it can only boot once and to find what it save that make it stop booting. late at night so I go to bed

[Béèm]

I have provided a significant amount of information on the problem of "Save-session disables the media from subsequent booting."

Questions for forum members

• If I make an ISO of the media, is there anyone in this forum who understands boot media well enough to review it for problems?
• If so, where should I post the ISO?
• Would it be acceptable to use "dd if=/mnt/sr0 of=wontboot.iso" be an acceptable method of presenting this for forum review?
• Should all the information provided in the other post be "zipped" into a package with the ISO so that its all in one place?
• Would anyone indicate whether this is acceptable?

Thanks in advance
P.S. Can anyone explain what makes a data media bootable for Linux OSs? Does the file location on CD have something to do with this or is it some other technology so that the CD/DVD device detects bootable media? (This might help in understanding how save-session is destroying the bootable live media.)

I have the impression you did post in the wrong thread.

[Béèm]

@ Béèm
I could end this thread with [Solved], but that is not really true.

It is solved for me, or I ran "around" the issue, but the "issue" still exists. It is not solved.

Unless you mean "solved for me".

I follow closely the forum and the issue you expose isn't seen before.
Therefor I think there is something particular in your Compaq machine.
So I don't think it can be considered as a general issue or flaw.

I agree it is a weird issue.

And the 'solution' to remaster is indeed uncommon.

I have no problems that you leave the thread open.
However I am not convinced that somebody can contribute more, but one never knows.

[gcmartin]

...

I have the impression you did post in the wrong thread.

Curiously, would anyone recommend where this, logically, should be posted or opened?

Thanks in advance.