Puppy Linux Discussion Forum Forum Index Puppy Linux Discussion Forum
Puppy HOME page : puppylinux.com
"THE" alternative forum : puppylinux.info
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

The time now is Sun 19 Nov 2017, 20:43
All times are UTC - 4
 Forum index » Advanced Topics » Puppy Derivatives
Easy Linux 0.5 pre-alpha, released Oct. 02, 2017
Moderators: Flash, JohnMurga
Post new topic   Reply to topic View previous topic :: View next topic
Page 20 of 23 [333 Posts]   Goto page: Previous 1, 2, 3, ..., 18, 19, 20, 21, 22, 23 Next
Author Message
technosaurus


Joined: 18 May 2008
Posts: 4756
Location: Kingwood, TX

PostPosted: Sat 14 Oct 2017, 21:34    Post subject:  

BarryK wrote:
BarryK wrote:

So, Easy does not run in RAM.

I also stated that this will be implemented properly in version 0.6, which means NOT YET.

In 0.5, the entire .session folder is wiped with "qfix=bak", which takes you back to a first-bootup situation.

So, no, you cannot bootup and then unplug the usb stick.

I could implement such a thing, but have no compelling reason to do so.


Rethinking this, it would probably be quite easy to implement. There are two possible ways:

1.
If "qfix=ram", copy working-partition into zram, and unmount working partition.
This has the advantage that do not have to remaster. At first bootup, set everything how you want, then add "qfix=ram" to kernel boot params, then reboot.

2.
Working partition is created in zram. So, will have a pristine new system at every bootup. You would have to remaster, to configure networking etc.
Or store various sized bz2'ed ext2 filesystems in the initramfs (they are really tiny when bz2 compressed) and bunzip the appropriately sized one to zram... This would give the option to save the session later - great for guest users. I don't know what happened to my patches that did this 'bunzip a compressed ext2 filesystem' trick went, but originally it went into the sfs directory so that settings established on first run were saved. It can always be resized or converted to ext3/4 later.
_________________
Check out my github repositories. I may eventually get around to updating my blogspot.
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Sat 14 Oct 2017, 21:48    Post subject:  

BarryK wrote:
BarryK wrote:

So, Easy does not run in RAM.

I also stated that this will be implemented properly in version 0.6, which means NOT YET.

In 0.5, the entire .session folder is wiped with "qfix=bak", which takes you back to a first-bootup situation.

So, no, you cannot bootup and then unplug the usb stick.

I could implement such a thing, but have no compelling reason to do so.


Rethinking this, it would probably be quite easy to implement. There are two possible ways:

1.
If "qfix=ram", copy working-partition into zram, and unmount working partition.
This has the advantage that do not have to remaster. At first bootup, set everything how you want, then add "qfix=ram" to kernel boot params, then reboot.

2.
Working partition is created in zram. So, will have a pristine new system at every bootup. You would have to remaster, to configure networking etc.


Well, I have implemented no.2:

http://bkhome.org/news/201710/running-easy-in-ram.html

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
belham2

Joined: 15 Aug 2016
Posts: 1301

PostPosted: Sun 15 Oct 2017, 04:29    Post subject:  

BarryK wrote:
BarryK wrote:
BarryK wrote:

So, Easy does not run in RAM.

I also stated that this will be implemented properly in version 0.6, which means NOT YET.

In 0.5, the entire .session folder is wiped with "qfix=bak", which takes you back to a first-bootup situation.

So, no, you cannot bootup and then unplug the usb stick.

I could implement such a thing, but have no compelling reason to do so.


Rethinking this, it would probably be quite easy to implement. There are two possible ways:

1.
If "qfix=ram", copy working-partition into zram, and unmount working partition.
This has the advantage that do not have to remaster. At first bootup, set everything how you want, then add "qfix=ram" to kernel boot params, then reboot.

2.
Working partition is created in zram. So, will have a pristine new system at every bootup. You would have to remaster, to configure networking etc.


Well, I have implemented no.2:

http://bkhome.org/news/201710/running-easy-in-ram.html



Will give this a try today.....thanks, Barry. Seriously though, if it makes it too hard and/or messes things up, just let it go. It's just I have ingrained in my head from using Precise all those years (when I did banking, etc online) that I always did it fully in RAM from a remastered Precise. Became 2nd nature, sort of.
Back to top
View user's profile Send private message 
lp-dolittle

Joined: 25 Apr 2016
Posts: 33

PostPosted: Sun 15 Oct 2017, 10:41    Post subject: keyboard layout setting
Subject description: xkbconfigmanager-1.2.pet
 

Barry, are you sure you are on the right scent with the xkbconfigmanager-1.2.pet?

When I install this PET in Easy OS 0.5 - or in quirky xerus64-8.3 - the XKB Configuration Manager no longer is usable (see attached screenshot).
At the same time (still on the Dell Latitude E6500 Laptop), the XKB Configuration Manager settings for the Swiss German Layout need some fine tuning, but remarkably SURVIVE restarts of X when I use an 'old' (2013) Puppy Precise 5.7.1 live CD!
On the Dell Latitude E6500 laptop the 'heavy' keyboard setting issue (spontaneous resets) only occurs in Quirky and Easy OS!

moreover:
The choice between the kernel modules i915 and gma500_cfx (as recommended by L18L) does not seem to influence the issue. I'm puzzled as to how manipulations of these kernel modules take effect. After I first believed that my changes had no effect, despite I had restarted X and rebooted,..... some restarts later - and after the laptop hang while booting (reporting: Intel 800/900 Series VBIOS HACK: version 0.5.3) - i915 was replaced by gma500_cfx.

could these symptoms indicate mutual interactions between several issues?
Screenshot xkblayout.png
Description 
png

 Download 
Filename  Screenshot xkblayout.png 
Filesize  122.87 KB 
Downloaded  34 Time(s) 
Back to top
View user's profile Send private message 
Rodney Byne

Joined: 31 Jan 2014
Posts: 149

PostPosted: Sun 15 Oct 2017, 18:17    Post subject: Why re-invent the wheel?  

Hi all,
instead of struggling to make complex Easy run in ram,
those of us with a longer memory shouldl just simply run
Quirky Werewolf compiled specifically for that purpose.

Suitable for online banking if needed and a smaller iso
than Easy to download as well.

Best regards.
Back to top
View user's profile Send private message 
belham2

Joined: 15 Aug 2016
Posts: 1301

PostPosted: Sun 15 Oct 2017, 18:50    Post subject: Re: Why re-invent the wheel?  

Rodney Byne wrote:
Hi all,
instead of struggling to make complex Easy run in ram,
those of us with a longer memory shouldl just simply run
Quirky Werewolf compiled specifically for that purpose.

Suitable for online banking if needed and a smaller iso
than Easy to download as well.

Best regards.



Hi Rodney,

Most likely my memory and use goes back way before yours, lol. Also, you are missing the forest for the trees Wink Barry has created something unique here with Containers. Having that, and running fully in RAM with an unplugged device that loaded everything???---you're talking about an unparalleled level of security and safety posture on the Net. That is the point here, not that we all---which most f us did and still do--- ran regular RAM for the past decade and a half, Quirky and all pups included.
Back to top
View user's profile Send private message 
technosaurus


Joined: 18 May 2008
Posts: 4756
Location: Kingwood, TX

PostPosted: Mon 16 Oct 2017, 00:28    Post subject:  

BarryK wrote:
Well, I have implemented no.2:

http://bkhome.org/news/201710/running-easy-in-ram.html
I am envisioning incognito mode: copy existing save file to (z)RAM ... keeps your settings but doesn't store anything permanently ... maybe call it guest mode
_________________
Check out my github repositories. I may eventually get around to updating my blogspot.
Back to top
View user's profile Send private message Visit poster's website 
scsijon

Joined: 23 May 2007
Posts: 1251
Location: the australian mallee

PostPosted: Tue 17 Oct 2017, 03:12    Post subject: w.t.h.!  


_________________
Puppy T290 - BarryK's Racy but with T2 Version 9.0. First Beta planned out end of August 2017.
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 03:47    Post subject: Re: keyboard layout setting
Subject description: xkbconfigmanager-1.2.pet
 

lp-dolittle wrote:
Barry, are you sure you are on the right scent with the xkbconfigmanager-1.2.pet?

When I install this PET in Easy OS 0.5 - or in quirky xerus64-8.3 - the XKB Configuration Manager no longer is usable (see attached screenshot).
At the same time (still on the Dell Latitude E6500 Laptop), the XKB Configuration Manager settings for the Swiss German Layout need some fine tuning, but remarkably SURVIVE restarts of X when I use an 'old' (2013) Puppy Precise 5.7.1 live CD!
On the Dell Latitude E6500 laptop the 'heavy' keyboard setting issue (spontaneous resets) only occurs in Quirky and Easy OS!


Do not use any xkbconfigmanager PET, it will not work. Nor will those files from npierce. They are designed to use /etc/X11/xorg.conf, whereas Easy uses /etc/X11/xorg.conf.d/10-evdev-puppy.conf.

In Quirky and Easy, xkbconfigurationmanager is builtin to woofQ, it is not a PET. It is different from the others out there.

Test it when 0.6 comes out.

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 03:59    Post subject:  

I have solved the problem of eth0 not connecting at first bootup.

0.5 did solve the problem of eth0 reconnecting after a reboot, but the first bootup remained unresolved.

I have fixed it, but puzzled. The fix was to insert "sleep 6" at the top of this script:

/etc/rc.d/rc.network_eth

This gets called, indirectly, from rc.sysinit at first bootup, and runs as a separate process, so bootup can continue.

What I found is that $INTERFACES was empty. "ifconfig -a" was only finding interface "lo". Putting in a sleep fixed it, and I was surprised just how much sleep was needed -- on my Mele mini-PC, it needs at least 4 seconds. I made it 6 to be on the safe side.

Then, the "ifplugstatus" tests are still required. Even with that prior "sleep 6", ifplug status has to run three times, that is another 2 seconds delay, before it reports an active network.

I don't know what has changed to require this sleep. Change in the kernel? Ethernet driver?

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 04:07    Post subject:  

Normally, before a new release, I update the DEB database, with the latest from Ubuntu "updates" repository.

In woof, this is done by running the "0setup" utility. This is the same thing that you can do in a running PKGget Package Manager.

I did this, then did a build, 0.5.90, a precursor to 0.6. Tested it on my Mele PCG35 mini-PC, but was getting awfully corrupted screen rendering. Artifacts left behind, parts of the screen not redrawing, or redrawing garbage, or redrawing parts of the screen in the wrong place.

This is most likely a problem with xorg and/or the intel driver.

It is disappointing to be going backwards. Over the years, have had rendering problems with the intel driver, but things have been pretty good for the last couple of years.

I think that screen rendering is pretty good with 0.5, though most of my usage has been with 0.4 and earlier.
So, am building Easy 0.6 with the Ubuntu DEBs as used in 0.5.

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 04:27    Post subject: Re: keyboard layout issue
Subject description: Intel graphics chipset
 

L18L wrote:
lp-dolittle wrote:
Details of the Dell LatitudeE6500 graphics controller can be seen from the attached screenshot

https://www.notebookcheck.net/Review-Dell-Latitude-E6500-Notebook.11958.0.html wrote:
The built in Intel Graphics Media Accelerator 4500M HD or a twice as fast and electrical power hungry dedicated NVIDIA Quadro NVS 160M graphic card takes care of graphic needs. The customer must again choose between low priced basic equipment and an upgrade, more efficient but for a surcharge . The built in 4500M HD graphic chip fulfills all daily demands, such as classic office usage, internet and video (even HD).
and
http://www.dell.com/support/home/de/de/debsdt1/Drivers/DriversDetails?driverid=R304345&lwp=rt wrote:
Intel GM45/GE45/GS45 Integrated Graphics Driver
This package provides the Intel GM45/GE45/GS45 Integrated Graphics Driver and is supported on Latitude E5400/E6400/E4200/XFR E6400/E6400 ATG/E6500/E4300/E5500 that are running the following Windows Operating Systems: Vista and Windows 7.
tell me that you should try to use kernel driver module gma500_gfx instead of i915.
Something like
Code:
insmod gma500_gfx && rmmod i915
and restartx.

Good luck!


Have you tried blacklisting each of these kernel modules?

Menu System -> Boot Manager

Then choose the button to blacklist modules.

Requires a reboot.

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 04:34    Post subject: Re: keyboard layout issue
Subject description: Intel graphics chipset
 

lp-dolittle wrote:
Moreover, the Swiss German layout obviously is only one among others that cannot be set easily. As reported by forum member ottod (Quirky Xerus 8.3 blog from 01.08.2017), the latam layout also does not work.


The post from ottod:

http://murga-linux.com/puppy/viewtopic.php?p=962726#962726

Quote:
For a long time Quirky has been lacking latam keyboard so initially I have to do with es keyboard, which is close enough. Afterwards, I select latam keyboard from the advanced Xorg keyboard configuration and everything works ok.
Right now in Quirky 8.3, this procedure works erratically. Some times it does execute the requested configuration and some times it stays setup to some variant of english, I suppose us or en-gb. I have not been able to identify the successful sequence. When I finally get the latam keyboard to work, it does not survive a reboot.
I'm using Quirky Xerus 8.3, USB flash stick image.


Yes, the keyboard selection in QuickSetup is for the terminal keyboard layouts. These are in /lib/keymapps, and there doesn't seem to be one corresponding to "latam".

So choosing "es", then "latam" in the Xkb Config Manager, does the trick. And that should be working ok for 0.6.

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
BarryK
Puppy Master


Joined: 09 May 2005
Posts: 8312
Location: Perth, Western Australia

PostPosted: Tue 17 Oct 2017, 06:05    Post subject: Re: keyboard layout issue
Subject description: Intel graphics chipset
 

BarryK wrote:
lp-dolittle wrote:
Moreover, the Swiss German layout obviously is only one among others that cannot be set easily. As reported by forum member ottod (Quirky Xerus 8.3 blog from 01.08.2017), the latam layout also does not work.


The post from ottod:

http://murga-linux.com/puppy/viewtopic.php?p=962726#962726

Quote:
For a long time Quirky has been lacking latam keyboard so initially I have to do with es keyboard, which is close enough. Afterwards, I select latam keyboard from the advanced Xorg keyboard configuration and everything works ok.
Right now in Quirky 8.3, this procedure works erratically. Some times it does execute the requested configuration and some times it stays setup to some variant of english, I suppose us or en-gb. I have not been able to identify the successful sequence. When I finally get the latam keyboard to work, it does not survive a reboot.
I'm using Quirky Xerus 8.3, USB flash stick image.


Yes, the keyboard selection in QuickSetup is for the terminal keyboard layouts. These are in /lib/keymapps, and there doesn't seem to be one corresponding to "latam".

So choosing "es", then "latam" in the Xkb Config Manager, does the trick. And that should be working ok for 0.6.


Have fixed Latin-American layout properly:

http://bkhome.org/news/201710/kmap-layout-for-latin-american.html

_________________
http://barryk.org/news/
Back to top
View user's profile Send private message Visit poster's website 
lp-dolittle

Joined: 25 Apr 2016
Posts: 33

PostPosted: Tue 17 Oct 2017, 16:50    Post subject: running Easy OS in RAM
Subject description: security
 

@ Belham2 and technosaurus

running an OS totally in RAM certainly provides advantages, not least in view of secure browsing. However, I cannot help questioning the limits of this enhanced security.
Please, try to reenact the following reasoning, and correct potential misapprehensions:

Running an OS totally in RAM, on a PC without internal hard-disk, can be considered save after the boot-medium (CD; DVD, SD-card, pen drive, etc) is removed, or maybe it preferably should be called 'as save as possible'? For sure, the original OS is in a safe place, but what about the currently running RAM version? How far is this transient installation vulnerable to malware attacks, while browsing the Internet? Is it conceivable that malware during this mode of operation - via invading the RAM - alters the functionality of the OS or simply settles down, in order to go viral in case a storage medium is attached and mounted? In fact, running an OS totally in RAM cannot spare users the necessity of, sooner or later, transferring data to a storage medium. Well, this storage medium, for good measure, next time can be mounted only after the OS again has been booted into RAM, but data may be corrupted! I don't think the 'cloud storage option' can be an alternative.
Finally, I also could imagine an additional security risk. What about the 'theoretical' storage options in the firmware of the computer's hardware components (EPROMs)? Can this kind of security risk be neglected/excluded?

Where is the line between absolute and maximum security?
Feel free to comment on the above-mentioned ideas and to criticise ... keeping in mind that it is absolutely not my intention to doubt the advantages of the 'running totally in RAM' option!

kind regards
Back to top
View user's profile Send private message 
Display posts from previous:   Sort by:   
Page 20 of 23 [333 Posts]   Goto page: Previous 1, 2, 3, ..., 18, 19, 20, 21, 22, 23 Next
Post new topic   Reply to topic View previous topic :: View next topic
 Forum index » Advanced Topics » Puppy Derivatives
Jump to:  

You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum
You cannot attach files in this forum
You can download files in this forum


Powered by phpBB © 2001, 2005 phpBB Group
[ Time: 0.1097s ][ Queries: 15 (0.0269s) ][ GZIP on ]