Dingo Multisession

Discuss anything specific to using Puppy on a multi-session disk
Post Reply
Message
Author
User avatar
Lobster
Official Crustacean
Posts: 15522
Joined: Wed 04 May 2005, 06:06
Location: Paradox Realm
Contact:

Dingo Multisession

#1 Post by Lobster »

My main Puppy for the past two weeks has been a Multisession of Puppy Dingo Alpha 6

Now Dingo is still in Alpha - not yet ready for general testing or usage . . .
. . . however the multisession has allowed me to run Puppy with no hard disk - just a DVD rewriter

One of the things I can not work out is where the devx_396.sfs goes on a multisession or indeed whether this can be done (devx_396.sfs is the file that runs the compiling programs in Puppy)
Normally devx_xxx.sfs are placed, if running from a CD and HD setup, in mnt/home/

The last time I used multisession :oops: was the first time Barry introduced it. It seemed incredible. An operating system on a CD and data saved on CD. 8)
I remember running with geek glee with my CD from one computer to another to see if a file saved on machine was still readable.
It was! This was of course a test version but it worked. 8)

Because of the inherent simplicity of Puppy, it is flexible enough to run in a variety of ways.

I know Flash is a user and advocate of this multisession method.

Suddenly we need Multisession and there it is. Just works.

Thanks guys :)
Puppy Raspup 8.2Final 8)
Puppy Links Page http://www.smokey01.com/bruceb/puppy.html :D

User avatar
Scoticus
Posts: 443
Joined: Mon 01 Jan 2007, 19:47
Location: Erskine, Scotland

Dingo Multisession

#2 Post by Scoticus »

When I was having difficulty with 2.16 and Open Office.sfs files loading to multisession BarryK advised that it was not possible.

With everything that has gone on from that time I would expect that it has been left on a to do list or at worst abandoned.

I note that a number of offerings come with .sfs files inbuilt to the ISO and this seems to get round the problem. You yourself have used this method in your 3.01 version.

Would this not do the trick? Perhaps you could encourage Barry to offer up a solution to loading .sfs to multisession disks which would once again get everyone back to the smaller downloads

jacknOOb
Posts: 3
Joined: Sun 17 Feb 2008, 15:18

#3 Post by jacknOOb »

Hi Lobster

Welcome back to multi-session. It's a whole new world!

I'm just new here myself and don't know much at all, but I think you'll find some clues in this post:
You're right. Multi-session is incredible. It's a shame some of its vital functionality dropped out some time ago with "improvements" made to the init scripts.
  • pfix=ram hasn't worked since puppy301
    pfix=<n> hasn't worked since puppy215CE
As much as I'd not want to see Dingo's development delayed, I'd hate to see these bugs perpetuated. These bugs of course don't affect all you hard drive installers, so you're probably not aware of them or how important they are to live-CD experimenters. I'd really appreciate you putting in a good word for us though. I hope it's not too hard to fix. I only wish I had the skills to fix it myself and I would.

Hope you can help.

User avatar
Flash
Official Dog Handler
Posts: 13071
Joined: Wed 04 May 2005, 16:04
Location: Arizona USA

#4 Post by Flash »

Lobster, I think you have to add the .sfs to the .iso, so that it appears in the root of your DVD. I think that's what I needed to do when I tried to use OpenOffice a long time ago. Turned out that I didn't have enough RAM or something, and I didn't have a hard disk drive at the time, so I gave up on it. If you're using a DVD+RW it's easy enough to try.

Jack, puppy pfix=2 worked on the Puppy 3.01 multisession DVD I'm using. I can see the (hidden) file (/.badfolders) it created, and when it boots, the multissession DVD skips the sessions I told it to. I haven't tried it in Dingo. I just discovered the other day that puppy pfix=ram doesn't work in Dingo; the multisession DVD still loads the sessions. Barry probably isn't aware of it. Yet. :wink:
[url=http://www.murga-linux.com/puppy/viewtopic.php?t=69321][color=blue]Puppy Help 101 - an interactive tutorial for Lupu 5.25[/color][/url]

jacknOOb
Posts: 3
Joined: Sun 17 Feb 2008, 15:18

#5 Post by jacknOOb »

Hi Flash

Of course, all this is a bit off-topic by now, but I'm just glad the issue has been brought to light. Please, do with it whatever you consider appropriate.

It seems to me that the puppy pfix=2 that you tried would not have worked if you only had 2 sessions saved, that is, you cannot pfix=n if you have exactly n folders saved. This entails the loss of what is conceivably a rather important function that would otherwise previously have been served by pfix=ram. Unfortunately, that no longer works either.

I am booting live and saving back to multi-session disk on DVD-R. There is no hard drive or flash drive involved nor has there been.

This machine is P4 2.8GHz 1GB with LG dual-layer burner. Most isos were burned as multi-session using BurnCDCC in Windows, some using burniso2cd in Puppy.

As near as I can tell from the versions presently available to me, this is the situation:

Following are the results of using the boot options on every Puppy I've seen rebooted after saving the initial configuration only, that is, when there is only one saved folder:
  • puppy-2.14-seamonkey-fulldrivers.iso.......................pfix=ram OK........pfix=<1> OK
    pizzapup-3.0.1.iso (based on Puppy 2.14)................pfix=ram OK........pfix=<1> OK
    teenpup2.14.iso........................................................pfix=ram OK........pfix=<1> OK
    puppy-215CE-Final.iso..............................................pfix=ram OK........pfix=<1> OK
    puppy-2.16.1-seamonkey-fulldrivers.iso...................pfix=ram OK........pfix=<1> fails
    puppy-2.17.1-nolzma-seamonkey-fulldrivers.iso.......pfix=ram OK........pfix=<1> fails
    tmxxine_prism (based on 2.17).................................pfix=ram OK........pfix=<1> fails
    (FireHYdrant)Inferno3.0b.iso.....................................pfix=ram fails.......pfix=<1> fails
    puppy-3.01-seamonkey.iso.......................................pfix=ram fails.......pfix=<1> fails
    puppy-3.01retro-k2.6.18.1-seamonkey.iso...............pfix=ram fails........pfix=<1> fails
    pcpuppyos_3.01-rc4.iso............................................pfix=ram fails........pfix=<1> fails
    tmxxine-301.iso........................................................pfix=ram fails........pfix=<1> fails
When pfix=ram fails, saved folders are loaded anyway instead of being ignored, no matter how many there are.
When pfix=<n> fails, the specified number of folders are successfully marked as bad but kernel panic ensues at a later stage of the boot sequence.

Further investigation has shown that pfix=1 only works if there are 2 or more saved folders, that is, you cannot mark all saved folders as bad, even if there's only one. If there are 5 saved folders, for example, pfix=5 fails while pfix=4, 3, 2 or 1 succeeds. Since pfix=ram no longer works as it did in earlier versions, there is consequently no convenient way to restore an installation to its original unconfigured state.

Of course, there may well be any number of workarounds, depending on what you are trying to achieve, but the fact remains that the features do not work as expected. Because they used to work in earlier versions, I would hope that this functionality could be restored fairly easily. If someone skilled in scripting could give some attention to this, I for one would very much appreciate it. I'm not interested in a workaround. In fact, I would really like to see this functionality of the multi-session Puppy not only restored but enhanced. Some very interesting possibilities come to mind in that direction .....

jacknOOb
Posts: 3
Joined: Sun 17 Feb 2008, 15:18

Dingo multisession

#6 Post by jacknOOb »

Hi all

Sorry to be a pain about this, but ...

Dingo 3.96 still had these pfix=ram and pfix=<n> bugs in multi-session.

Thankfully, someone (Flash?) alerted Barry to this thread and he found the trouble with pfix=ram, as he reported in his blog March 6: http://puppylinux.com/blog/?cat=1&paged=3. When he released Dingo Alpha7 the next day, it was fixed.

However, pfix=<n> still doesn't work as expected, as it used to before Puppy216.

I now have three reports, from Oli, Flash and Barry, all insisting that it does work. Apparently, they haven't tested it in any systematic fashion. The bug can easily be reproduced by counting the number of previously saved folders on the multisession CD/DVD, rebooting without saving and entering the boot option puppy pfix=<n>, where n is equal to that count. The specified number of folders are successfully marked as bad but kernel panic ensues. So it's not that pfix=<n> doesn't work. Mostly it does. It's just that pfix=<all> doesn't work.

This is no big deal. Since pfix=ram is now working again, I suppose that it's analogous to pfix=<all>. But the fact remains that pfix=<n> still doesn't work as expected. This is exactly what I encountered inside an hour or so of first using Puppy. What bugs me is that it's something on the very first boot screen that has the potential to result in kernel panic.

Now I'm trying to introduce Puppy to a few people and I'm just hoping his debut isn't marred by his piddling on their shoes. Some of these people are quite discerning, okay, picky. Not all of them are likely to try what I tried, but the focus of my project entails the use of the multi-session disk. Actually, my eventual objective is to come up with something that is idiot-proof and bomb-proof, and Puppy is certainly the closest thing I've seen to that. I'm talking kids and seniors here.

One other thing I should point out is that, when using the pfix=<n> boot option, these folders are not necessarily "permanently marked bad", if the user elects NOT to save the session. So, unless you want to try something fancy, there's ordinarily no need to muck with /.badfolders.

Thanks to Barry and everyone and please excuse my petty foibles.

User avatar
Flash
Official Dog Handler
Posts: 13071
Joined: Wed 04 May 2005, 16:04
Location: Arizona USA

#7 Post by Flash »

Jack, you're right that I haven't tested pfix=<n> systematically. That won't keep me from contributing some random thoughts and observations will it? :lol:

I think that you must create a first session to make a multisession Puppy, even if you don't explicitly save any settings when you shut down the first time. I may have that a bit garbled; it's been several months since I burned a multisession disk. Anyway, if you mount a multisession disk you can investigate the first saved session to see what's in it. If that first session is essential, perhaps that's why Puppy panics when you tell it to ignore it. I suppose it might be instructive to create a multisession disk but without otherwise saving any settings, then try to boot with pfix=1. I might try it, if the forum speeds up a bit and I can catch up with everything else I've got to do.

By the way, I didn't send Barry anything to inform him that pfix=ram didn't work. I get the impression that he reads the forum pretty closely. Obviously, he can't take the time to reply to everything he reads. :)

User avatar
BarryK
Puppy Master
Posts: 9392
Joined: Mon 09 May 2005, 09:23
Location: Perth, Western Australia
Contact:

#8 Post by BarryK »

In the case of the 'devx' file, I did consider that too, for alpha7. Um, now, what did I do? Did I mention it in the blog? If I recall rightly, if you download the devx_397.sfs file to '/' then reboot, if you have enough ram it will get loaded into ram then mounted as a unionfs layer.
[url]https://bkhome.org/news/[/url]

Post Reply