Puppy Linux Windows Installer - LICK v1.3.3 released

A home for all kinds of Puppy related projects
Message
Author
User avatar
ally
Posts: 1957
Joined: Sat 19 May 2012, 19:29
Location: lincoln, uk
Contact:

#346 Post by ally »

norby, where have you been!

over 2750 puppy iso's up, copies of ibilio repositories and pets, sfs, delta's and more for many others

somewhere around the 1Tb mark up, more going up everyday

:)

User avatar
Chili Dog
Posts: 161
Joined: Tue 20 Dec 2011, 11:17

#347 Post by Chili Dog »

Using the installer on 8.1 gives "unknown Windows version". And, judging by earlier posts about Windows 8 not booting after use, the Puppy Linux Windows Installer is obviously obsolete, and dangerous on modern computers.

F*ck Microsoft for what they have done to our machines, and hurting Linux in this way.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#348 Post by mikeb »

the Puppy Linux Windows Installer is obviously obsolete, and dangerous on modern computers.
a bit strong mate...I believe 'new and unventured territoty ' would be better.... norby has done the trick for all other versions...why not this one.

Actually what you meant to say was that windows was obsolete and the sooner the better :D

On the other side yer average window 8.1 machine is going to be a PITA on linux anyway...why not use the glorious OS you paid for until something better comes along?


mike

User avatar
Chili Dog
Posts: 161
Joined: Tue 20 Dec 2011, 11:17

#349 Post by Chili Dog »

No it's not, because, as others have noted, changes in Windows 8 boot method mean that using this app could break your Windows 8 installation. Therefore, it's dangerous.

Why don't you go use the Windows you paid for? You know, since it comes on every PC sold in stores whether you like it or not. (Enjoy your Microsoft tax.)

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#350 Post by mikeb »

Simply a suitable method of using this instaler has yet to be formulated.
I assume its this that you are talking about and not problems from repartitioning or UFEI

For vista and 7 (and possibly 8) I just rename bootmgr and replace with grldr and use chainloading...I wonder how that would work with 8.1?

Use the windows you paid for...after a year or 3 Linux distros will have built up driver support and then use it and dump the windows.

Perhaps you could do some testing if you have a machine with 8.1 on it.

mike

noryb009
Posts: 634
Joined: Sat 20 Mar 2010, 22:28

#351 Post by noryb009 »

Chili Dog: Windows 8.1 support is currently in the repository, I just haven't gotten around to recompiling every installer and uploading them. If you need 8.1 support, you can compile it yourself, there's only a few steps (but I'll probably be recompiling the latest later today, I didn't realize it didn't have 8.1). I'm working on another project to remedy this, but I haven't had much time to work on it.

Windows 8.1 is tricky, since almost all 8.1 machines use UEFI, and grub4dos doesn't support UEFI. To make the installer work, it needs to replace the windows bootloader with a UEFI-friendly linux one, then chainload either windows or a version of puppy. I don't want to implement this in NSIS, so I'm including this change in the new project.

Your jumping to conclusions about the program being dangerous - IIRC, I've seen one post where data was lost using puppy+windows 8, and I don't know if they even used the installer. Meanwhile, there have been thousands of downloads of the installer. If it was actually dangerous, we would hear of a lot more problems.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#352 Post by mikeb »

I was wondering that...using gparted on windows 8.1 I would expect to have problems for example.

mike

xcleaner
Posts: 11
Joined: Mon 21 Dec 2009, 10:52

boot error 27 in grub4dos

#353 Post by xcleaner »

I installed LXpup 14.12 with NSI with no trouble on win7
but when reboot i choose start linux
and it start grub4os with this error :find --set-root --ignore-floppies /LXpup 14.12/initrd.gz error 27 Unrecognised command...
i don't know what to do...can you help me!
here my menu.lst

timeout=5
default=0
gfxmenu=/Grub_GUI.gz

title LXpup 14.12 520
find --set-root --ignore-floppies /LXpup 14.12/initrd.gz
kernel /LXpup 14.12/vmlinuz psubdir="LXpup 14.12" pfix=fsck
initrd /LXpup 14.12/initrd.gz
boot

title LXpup 14.12 520 - Do not load save file
find --set-root --ignore-floppies /LXpup 14.12/initrd.gz
kernel /LXpup 14.12/vmlinuz psubdir="LXpup 14.12" pfix=ram
initrd /LXpup 14.12/initrd.gz
boot


title LXpup 14.12 520
find --set-root --ignore-floppies /LXpup 14.12/initrd.gz
kernel /LXpup 14.12/vmlinuz psubdir="LXpup 14.12" pfix=fsck
initrd /LXpup 14.12/initrd.gz
boot

title LXpup 14.12 520 - Do not load save file
find --set-root --ignore-floppies /LXpup 14.12/initrd.gz
kernel /LXpup 14.12/vmlinuz psubdir="LXpup 14.12" pfix=ram
initrd /LXpup 14.12/initrd.gz
boot

I'am in triple boot with windows 7 and 8.1
before the installation of Lxpup on win 7 i used EASYBCD

noryb009
Posts: 634
Joined: Sat 20 Mar 2010, 22:28

#354 Post by noryb009 »

Try using this as your menu.lst:

Code: Select all

timeout=5
default=0
gfxmenu=/Grub_GUI.gz

title LXpup 14.12 520
find --set-root --ignore-floppies "/LXpup 14.12/initrd.gz"
kernel "/LXpup 14.12/vmlinuz" psubdir="LXpup 14.12" pfix=fsck
initrd "/LXpup 14.12/initrd.gz"
boot

title LXpup 14.12 520 - Do not load save file
find --set-root --ignore-floppies "/LXpup 14.12/initrd.gz"
kernel "/LXpup 14.12/vmlinuz" psubdir="LXpup 14.12" pfix=ram
initrd "/LXpup 14.12/initrd.gz"
boot 
Did you compile this yourself? If so, in the future, don't use spaces in the install-dir, use dashes or underscores instead

xcleaner
Posts: 11
Joined: Mon 21 Dec 2009, 10:52

boot error grub4dos error 27

#355 Post by xcleaner »

Thank for your quick answer i use your menu.lst
but same issue ...error 27 unreconized command
I compile it with nsi mainkensisw
may be it's comming from EASYDCD menu..
here he is

Il y a 3 entrées dans le chargeur d'amorçage de Windows

Entrée par défaut : Windows 8.1
Compte à rebours : 5 seconde(s)
Disque d'amorçage : D:\

Entrée #1
Nom : Windows 8.1
Identifiant BCD : {default}
Disque : G:\
Chemin du chargeur d'amorçage : \Windows\system32\winload.exe

Entrée #2
Nom : Windows 7
Identifiant BCD : {current}
Disque : C:\
Chemin du chargeur d'amorçage : \Windows\system32\winload.exe

Entrée #3
Nom : Start Puppy Linux
Identifiant BCD : {62e34ff6-7229-11e4-90ad-ad3c569c2a53}
Disque : C:\
Chemin du chargeur d'amorçage : \grldr.mbr

An other suggestion ?? thank's

xcleaner
Posts: 11
Joined: Mon 21 Dec 2009, 10:52

error boot gru4dos error27

#356 Post by xcleaner »

You was right....
i just uninstall and then compile again without spaces..
And everything is ok now..i can normaly start my LXpup 14.12

Thank's a lot for you help!

noryb009
Posts: 634
Joined: Sat 20 Mar 2010, 22:28

#357 Post by noryb009 »

Good to hear. Thank you, posting the menu.lst certainly made the debugging easier! :lol:

Snail
Posts: 331
Joined: Sun 18 Oct 2009, 07:32

Odd behaviour of Slacko 5.7 installed with the Installer

#358 Post by Snail »

Thank you very much for this hugely positive development in Puppy Noryb.

I have run into a very odd behaviour of Slacko after installing Slacko 5.7 onto an HP Mini netbook. The problem only occurs with the non-PAE version. As far as I have been able to tell, the PAE version works fine.

The machine specs are:

HP Mini 110-3500, CPU is Intel Atom N550, dual core(4 thread), 1.5GHz processor incorporating GMA3150 Graphics Accelerator. The pre-installed OS is Win 7 Starter.

As there is only 1G of RAM, I assumed that the non-PAE version would be appropriate. This may be wrong.


I installed slacko non-PAE 5.7, using the installer exe file from your website (MD5sum checked OK).

The installation went well and I was able to boot Slacko the first time, in about 1 minute. (Of that 1 minute, the last 20 seconds or so were following the message "Slacko Puppy Linux/Linux 3.4.82[i686 arch]. Since this message appears when X is restarted, I assume that the last 20 seconds are all involved with starting X and loading the desktop.)

I did not make any changes in the first boot customisation setup panel, nor attempt to access the Internet. I then saved and exited. On reboot, everything looked good, until X started loading. This step then took over 6 minutes, compared with the 20 seconds on first boot! Most unfortunate. (When the desktop finally became available, Slacko seemed reasonably responsive).

Restarting X has the same problem. Running Xorgwizard, using all the recommended options, didn't help and of course it finished with a 6 minute restart of X.

I tested repeatedly, with two different save files - same problem every time.

Booting pfix=ram there is no problem.

It seems as though the initrid can detect and handle the hardware perfectly on first boot but, instead of helping, the savefile greatly hinders things on subsequent boots, in spite of the fact that I follow the defaults.

As the problem does not occur with the PAE version of the Slacko 5.7 installer on your website, I am good to go. I report this in case you find the problem is interesting!

I don't know if the problem would also occur if I installed Slacko the hard way, so I'm unsure if the problem is related to Slacko non-PAE itself or just the Installer version. I will link to this post on the Slacko site, in case miko finds it interesting as well.

It would be great if there was any chance of a fast boot/secure boot/UEFI compatible version of the Installer. James Bond and Kirk have cracked the problem with Fatdog.

User avatar
rcrsn51
Posts: 13096
Joined: Tue 05 Sep 2006, 13:50
Location: Stratford, Ontario

#359 Post by rcrsn51 »

Is your Puppy install in the Windows NTFS partition?

When you created the savefile, what filesystem type did you pick? I recall a similar situation where ext2 was MUCH faster than other formats.

noryb009
Posts: 634
Joined: Sat 20 Mar 2010, 22:28

#360 Post by noryb009 »

Your problem is strange, being PAE or not shouldn't change the way X starts, and the installers were created the same way. Best I can recommend is try defragmenting the drive the save file is on. It doesn't sound like it is caused directly from the installer, so you will probably get more help from others.

I'm slowly but surely making headway in rewriting the installer, which will lead to supporting UEFI/secure boot (but not fast boot, as it's only useful it you aren't dual booting).

Snail
Posts: 331
Joined: Sun 18 Oct 2009, 07:32

PAE Weirdness

#361 Post by Snail »

I installed Slacko into sda2, the Win 7 NTFS C: partition. This partition is 216GB and is only half full. It was defragmented the day before installation and first boot of slacko. Nothing much was done to it between defrag and install.

The savefile is ext4, non-encrypted. However, there does not seem to be a general problem of slowness, the non-PAE seemed about as quick as the PAE install, with the exception of the X-factor! Wouldn't any slowness associated with a journalled filesystem be seen during writing rather than reading? And wouldn't the X step of the boot be largely a read operation?

Snail
Posts: 331
Joined: Sun 18 Oct 2009, 07:32

rcrsn51 was right!

#362 Post by Snail »

rcrsn51 is a genius! And I'm just a snail.

I re-installed slacko non-PAE and made 3 pupsaves, an ext2, an ext3 and an ext4. All were made without changing any of the first boot defaults and without attempting to connect to a network. They are all in the default folder, Fido is not used, they are unencrypted and they are all 512MB.

With the ext2 savefile the Xstep in boot up takes less than 20 seconds, ext3 takes about 5minutes 20secs and ext4 just over 6 minutes.

By comparison, the X step with an ext4 savefile that has been customized a bit takes 13 seconds with the PAE version of slacko.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#363 Post by mikeb »

Well wild differences...I would be reporting this elsewhere since it may be of significance to others and as mentioned out of th domain of the windows installer.

Only though that popped in my head was could this be related to journalling and puppies dirty / unclean shutdown behaviour....
Though ext4 for saves it journalled or not?? seems to vary.

How big were the save files.... curious how PAE handling was affecting this...sounds more like a kernel config problem between to two....indeed that may also account for the save format variations.....something dmesg output may reveal.
When I used to use save files I never noticed a speed difference between ext2 and ext3..only that ext2 corrupted quickly so this to me is not expected behaviour and if it is now something is fundamentally broken.

mike

User avatar
rcrsn51
Posts: 13096
Joined: Tue 05 Sep 2006, 13:50
Location: Stratford, Ontario

#364 Post by rcrsn51 »

This is not a PAE vs nonPAE issue. It has to do with the version of the ntfs-3g driver that was used in Slacko57 nonPAE. Because your savefile is in an NTFS partition, ntfs-3g is in play, and it bogs down when working with journaled filesystems.

User avatar
mikeb
Posts: 11297
Joined: Thu 23 Nov 2006, 13:56

#365 Post by mikeb »

Ah ok.... so an older version of ntfs-3g might work better? or newer.....
just for general reference. Is there a range of puppy releases this would affect or just the slackos...?

Make sense since I don't think I ever used save files on ntfs only fat which is ironic since barry was told it was wrong to have a journaled image filesystem on a non journaled partition..... ext3 on fat worked just fine.

mike

Post Reply