colinux-users — coLinux users mailing list

Re: [coLinux-users] AMD64 cooperative.h

[Henry N. <Hen...@Ar...>]

Shannon Barber wrote:
> I checked-out the tip of the stable colinux with svn and am attempting 
> to build a kernel that will work with my Gentoo install.
> http://colinux.svn.sourceforge.net/svnroot/colinux/branches/stable (rev 980)
> 
> Is there a asm-x86_64/cooperative.h & asm-x86_64/cooperative_internal.h ?
> I just copied the ones fro i386 and its building now, but I'm concern 
> about it crashing/misbehaving at run-time.
> 
> Has anyone already done this? Or is work needed to create proper x86_64 
> versions of these files?

64bit kernel don't exist and needs more changes as copy files only.

Please follow steps "4. Building a coLinux kernel" in file doc/building
http://colinux.svn.sourceforge.net/svnroot/colinux/branches/devel/doc/building

-- 
Henry N.

[coLinux-users] AMD64 cooperative.h

[Shannon B. <sgb...@wo...>]

I checked-out the tip of the stable colinux with svn and am attempting 
to build a kernel that will work with my Gentoo install.
http://colinux.svn.sourceforge.net/svnroot/colinux/branches/stable (rev 980)

Is there a asm-x86_64/cooperative.h & asm-x86_64/cooperative_internal.h ?
I just copied the ones fro i386 and its building now, but I'm concern 
about it crashing/misbehaving at run-time.

Has anyone already done this? Or is work needed to create proper x86_64 
versions of these files?

Re: [coLinux-users] coLinux-users Digest, Vol 23, Issue 16

[<Use...@zo...>]

dis...@bo...(insomniux)  20.03.08 22:28


>>> Have a look on "IPcop".
>>> It should work under colinux too and there was "once" (1 year or
>>> more ago) in heise c't an article howto to integrate IPcop into a
>>> VM in Windows.
>>> I am quite sure that would be the easiest way to avoid reinventing
>>> the wheel ;-)
>Good suggestion. IPcop seems to be faster but I don;t know if their
>content filtering algorithm is as good as dansguardian?

I looked at IPcop:
Currently (1.4.18) it is running linux 2.4 so assume it would
not work under the colinu 2.6 kernel

OTHO:
An attempt to use colinux as firewall is made here:
http://colinux.wikia.com/wiki/UsingCoLinuxAsAFirewall

maybe harden, bastille, smothwall, or so makes configuring iptables easier.




Rainer

Re: [coLinux-users] giving a uuid in config file / command line

[<Use...@zo...>]

and...@gm...(Andrew Roth)  21.03.08 11:29


>It does have a drive letter in windows, but I haven't had any problems
>with it.  

...yet.

Lucky you!

Windows is noting the time of last access, so you can't
avoid Windows writing to the directory structure, for example if
the virusscanner is working. Linux is noting that access too.

What will happen if Colinux and Windows
are writing to the same directory structure part?

So "only reading" would not help, as the time stamps do still "write".

Mount it "read only" to avoid linux updating access time or turn time
stamping off.

BTW:
Does anybody know how to mount a partition "read only" under Windows?



Rainer---<=====>                         Vertraulich
             //                          Key-ID:38F34C59  
           //                            
         <=====>--------------ocholl, Kiel, Germany ------------

Re: [coLinux-users] giving a uuid in config file / command line

[Henry N. <Hen...@Ar...>]

Hello Andrew,

if you have a drive letter under Windows, then Windows has opened the
drive and not all buffer must be in the dist. Linux with cobd opens the
same drive but not known about non written file buffers from windows
side. So you can got some other datas from file in inside coLinx as on
the Windows side. If you would create directories on Windows side,
coLinux would not see it, or some times later, after windows was writte
the buffers to disk.

The fsync on cofs was fixed. Please use a newer version from
http://www.colinux.org/snapshots/

Cofs has no UUID, cofs is not a disk it is a filesystem (same as nfs,
smbfs).

You means to deal with dualboot?
I have hack for SuSE 9.0. There I have changed the exclude list for
"mount -a" to have cofs under coLinux and nfts under native boot. On
coLinux I mount the cofs to the same place with the '--bind' option.

This are the lines in /etc/fstab:
/dev/hda2  /media/WinXP  ntfs
ro,auto,uid=hn,gid=users,user,fmask=0113,dmask=0002 0 0
cofs0:/  /media/cofs0  cofs
auto,user,noexec,uid=hn,gid=users,fmask=0664,dmask=0775 0 0


Changes and adds on /etc/init.d/boot.localfs:

... near the top I have ...
# Switch between coLinux and real linux
if uname -r | grep -qe "-co-"
then
        COLINUX=true

        echo " FSCK skip (colinux)"
        export fastboot=true

        # Mounts for coLinux
        COLINUX_EXCLUSE="nontfs,novfat"
else
        COLINUX=false

        # Mounts for native Linux
        COLINUX_EXCLUSE="nocofs"
fi

... some lines later ...

#mount -av -t nonfs,noproc,nodevpts,nosmbfs
mount -av -t nonfs,noproc,nodevpts,nosmbfs,$COLINUX_EXCLUSE

# Remount from cofs into original place for ntfs
if $COLINUX ; then
        mount --bind /media/cofs0       /media/WinXP
fi



Andrew Roth wrote:
> It does have a drive letter in windows, but I haven't had any problems
> with it.  I would use cofs, but if I can be blunt, I found it pretty
> useless because I got "Fsync failed" when I tried to do anything
> useful.  For example, I got that error when I tried to edit a file in
> vim, and also got it when I tried to run /script/server in a ruby
> application, which runs a light web server to test a ruby on rails
> application.  But I guess it's useful for straight copies.
> 
> Ok then, can we give cofs a uuid?  That way I can use the same
> /etc/fstab file which specifies the uuid now instead of a /dev/.. path.
> 
> -Andrew
> 
> On Fri, Mar 21, 2008 at 8:23 AM, Henry Nestler <Hen...@ar...
> <mailto:Hen...@ar...>> wrote:
> 
>     Andrew Roth wrote:
>     > Hi all,
>     >
>     > Is it possible to give a uuid on the config / command line in a
>     similar
>     > way that you can give an eth device a MAC address?
>     >
>     > ex
>     >
>     > hda3=\Device\Harddisk0\Partition3, 94749C99749C7FA2 # <-- specify uuid
>     >
>     > where the line in fstab is
>     >
>     > # /dev/sda3
>     > UUID=94749C99749C7FA2 /media/sda3     ntfs  
>      defaults,umask=007,gid=46 0
>     >
>     > Because for some reason my ntfs partitions get a different uuid in
>     > colinux.  Actually, I would be interested to know if that would be
>     a bug
>     > or not.
> 
>     The UUID is reading from filesystem, are some bytes from filesystem
>     superblock and is named GUID of volume.
> 
>     Big warning! Don't mount a ntfs partition, if it is online in Windows
>     (have a drive letter). You risk reading wrong datas.
> 
>     You should use cofs for accessing this drive.
> 
>     --
>     Henry N.
> 
> 

-- 
Henry N.

Re: [coLinux-users] giving a uuid in config file / command line

[Andrew R. <and...@gm...>]

It does have a drive letter in windows, but I haven't had any problems with
it.  I would use cofs, but if I can be blunt, I found it pretty useless
because I got "Fsync failed" when I tried to do anything useful.  For
example, I got that error when I tried to edit a file in vim, and also got
it when I tried to run /script/server in a ruby application, which runs a
light web server to test a ruby on rails application.  But I guess it's
useful for straight copies.

Ok then, can we give cofs a uuid?  That way I can use the same /etc/fstab
file which specifies the uuid now instead of a /dev/.. path.

-Andrew

On Fri, Mar 21, 2008 at 8:23 AM, Henry Nestler <Hen...@ar...> wrote:

> Andrew Roth wrote:
> > Hi all,
> >
> > Is it possible to give a uuid on the config / command line in a similar
> > way that you can give an eth device a MAC address?
> >
> > ex
> >
> > hda3=\Device\Harddisk0\Partition3, 94749C99749C7FA2 # <-- specify uuid
> >
> > where the line in fstab is
> >
> > # /dev/sda3
> > UUID=94749C99749C7FA2 /media/sda3     ntfs    defaults,umask=007,gid=46
> 0
> >
> > Because for some reason my ntfs partitions get a different uuid in
> > colinux.  Actually, I would be interested to know if that would be a bug
> > or not.
>
> The UUID is reading from filesystem, are some bytes from filesystem
> superblock and is named GUID of volume.
>
> Big warning! Don't mount a ntfs partition, if it is online in Windows
> (have a drive letter). You risk reading wrong datas.
>
> You should use cofs for accessing this drive.
>
> --
> Henry N.
>

Re: [coLinux-users] giving a uuid in config file / command line

[Henry N. <Hen...@Ar...>]

Andrew Roth wrote:
> Hi all,
> 
> Is it possible to give a uuid on the config / command line in a similar
> way that you can give an eth device a MAC address?
> 
> ex
> 
> hda3=\Device\Harddisk0\Partition3, 94749C99749C7FA2 # <-- specify uuid
> 
> where the line in fstab is
> 
> # /dev/sda3
> UUID=94749C99749C7FA2 /media/sda3     ntfs    defaults,umask=007,gid=46 0
> 
> Because for some reason my ntfs partitions get a different uuid in
> colinux.  Actually, I would be interested to know if that would be a bug
> or not.

The UUID is reading from filesystem, are some bytes from filesystem
superblock and is named GUID of volume.

Big warning! Don't mount a ntfs partition, if it is online in Windows
(have a drive letter). You risk reading wrong datas.

You should use cofs for accessing this drive.

-- 
Henry N.

[coLinux-users] giving a uuid in config file / command line

[Andrew R. <and...@gm...>]

Hi all,

Is it possible to give a uuid on the config / command line in a similar way
that you can give an eth device a MAC address?

ex

hda3=\Device\Harddisk0\Partition3, 94749C99749C7FA2 # <-- specify uuid

where the line in fstab is

# /dev/sda3
UUID=94749C99749C7FA2 /media/sda3     ntfs    defaults,umask=007,gid=46 0

Because for some reason my ntfs partitions get a different uuid in colinux.
Actually, I would be interested to know if that would be a bug or not.

thanks
-Andrew

Re: [coLinux-users] coLinux-users Digest, Vol 23, Issue 16

[<Use...@zo...>]

dis...@bo...(insomniux)  20.03.08 22:28

>theoretically education would be the best way.

I think parents sitting together with the children would be the best, 
wouldn't it? ;-) You know children (usually boys) starting early with TV, 
games etc. become typically not so smart tha other (mostly girls) because 
the "input" they got from that is only a fraction of what childrens brain 
was made for and needs. 
But it's not the place to discus here ;-)


>I've got 1G, hope that's enough

Have a lookmin task manager "performance" phyiscal RAM free"

>>>>My winXP will only use one of the cores (I think).
>>>
>>> XP would use "SMP" with upto 2 CPU(core)s.
>>> Have a look in taskmanager: You should see both cores there.
>>> If not, you run the wrong kernel. Test MS-KB for more info.
>You're right. Didn't look at the right place. It does use both cores.

>>> Have a look on "IPcop".
>>> It should work under colinux too and there was "once" (1 year or
>>> more ago) in heise c't an article howto to integrate IPcop into a
>>> VM in Windows.
>>> I am quite sure that would be the easiest way to avoid reinventing
>>> the wheel ;-)
>Good suggestion. IPcop seems to be faster but I don;t know if their
>content filtering algorithm is as good as dansguardian?

dansguarian, squid, mail filter, time manager, block out going traffic 
etc. are all made as essy to maintain "plugins/adons".
http://www.copfilter.org/
http://www.mhaddons.tk/
http://www.h-loit.de/index.php?option=com_weblinks&catid=23&Itemid=23
http://www.ipcop.org/index.php?module=pnWikka&tag=IPCopAddons
http://www.dageek.co.uk/ipcop/addonz/dansguardian.htm


If you not only want to "automatically protect" your children, but protect
your LAN/PC too i would recommand the "BOT"-addon: 
So all traffic MUST go thru the proxy. No P2P possible, no worm
can sent its spam, nor DNS can be manipulted by a broken hosts on the client.
But that's becoming to detailed for this list.


I currently don't know if one ever made an image of IPcop for coLinux.
and i don't know if the kernel would work.



Rainer

Re: [coLinux-users] DamnSmallLinux on colinux

[Henry N. <Hen...@Ar...>]

insomniux wrote:
> Meanwhile I've tried to setup archlinux, but for some reason I cannot
> use pacman to install anything. On every soruce I get an 'directory not
> found' or 'file not found', or so. Still the internet connection does work.
> E.g. in the the Debian fs, I can use apt-get to install apps. Any
> suggestions?

You use coLinux ArchLinux image? Or self made from scratch?

Here is a ready to use image:
http://sourceforge.net/project/shownotes.php?release_id=436233&group_id=98788

This is a step by step howto create a new ArchLinux:
http://wiki.archlinux.org/index.php/CoLinux

Some older docs, I'm afraid outdated:
http://colinux.wikia.com/wiki/Howto:_ArchLinux_install_from_iso_with_rescue_fs

Hope, you will find sugestions there.

-- 
Henry N.

Re: [coLinux-users] DamnSmallLinux on colinux

[insomniux <dis...@bo...>]

> >  
> > > =====Windows-console-output=========
> > > D:\Mike\colinux\dsl>C:\progra~1\colinux\colinux-daemon @dsl01.conf
> > > cofs0=d:\mik
> > > \colinux\dsl root=/dev/ram0 -t nt #ramdisk_size=40000
> > > Cooperative Linux Daemon, 0.7.2
> > > Compiled on Jan 18 2008 21:47:09
> > > 
> > > ...
> > > Kernel command line: root=/dev/hda1 ro root=/dev/ram0
> > > #ramdisk_size=40000
> > 
> > Typed wrong or you wand not use "ramdisk_size=..." ?
No, it's commented out (at least that's what I meant to do)

> > > ...
> > > RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
> > 
> > 4096 blocks a 1024 was set from coLinux kernel config:
> > CONFIG_BLK_DEV_RAM_SIZE=4096
> > CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024
> > 
> > > ...
> > > RAMDISK: Compressed image found at block 0
> > > RAMDISK: incomplete write (-28 != 32768) 4194304
> > 
> > The initial error: Write to ramdisk failed.
> > 
> > > VFS: Mounted root (ext2 filesystem) readonly.
> > > Freeing unused kernel memory: 128k freed
> > > attempt to access beyond end of device
> > > ram0: rw=0, want=32776, limit=8192
> > 
> > Your initrd size is to small. You have try to increase the ram disk
> > size? That should help.
> > What should do the '#' there?
> > "#ramdisk_size=40000" --> "ramdisk_size=40000"
> > 
> > Perhaps more ideas for you on Colonist?
> > I found in old user mail (2005-02-06)
> > http://sourceforge.net/mailarchive/message.php?msg_id=cu4pt5%24872%241%40sea.gmane.org
> > 
I've found Colonist, but did not find any documentation. Therefor it's
a bit hard to understand how to use it. 

Meanwhile I've tried to setup archlinux, but for some reason I cannot
use pacman to install anything. On every soruce I get an 'directory not
found' or 'file not found', or so. Still the internet connection does work.
E.g. in the the Debian fs, I can use apt-get to install apps. Any
suggestions?

Re: [coLinux-users] coLinux-users Digest, Vol 23, Issue 16

[insomniux <dis...@bo...>]

> > >Hi,
> > >For my children I have a dual boot machine with Linux and WIndowsXP.
> > 
> > >In Linux I use Dansguardian to keep the bad world out, 
> > 
> > Solving social problems with techinical matters never worked!

> > Educate your children, give them "media competence", surf together with them, 
> > but don't restrict them as dummies!
> > But ok, that's "your" children...
Sounds educated. Do you lock your car when you leave it at the railway
station? I do because I don't want to keep around all the time. On the
net may be confronted with the most terrible matter quite unrequested. I
do protect my children until they are able to cross the street by their
own. But still you're right, theoretically education would be the best
way.

> > >2. How much will this approach affect the performance of the PC
> > 
> > If you have lot of memory you would not notice it while surfing.
> > 
I've got 1G, hope that's enough

> > >My winXP will only use one of the cores (I think). 
> > 
> > XP would use "SMP" with upto 2 CPU(core)s.
> > Have a look in taskmanager: You should see both cores there.
> > If not, you run the wrong kernel. Test MS-KB for more info.
You're right. Didn't look at the right place. It does use both cores.

> > Have a look on "IPcop".
> > It should work under colinux too and there was "once" (1 year or
> > more ago) in heise c't an article howto to integrate IPcop into a 
> > VM in Windows.
> > I am quite sure that would be the easiest way to avoid reinventing 
> > the wheel ;-)
Good suggestion. IPcop seems to be faster but I don;t know if their
content filtering algorithm is as good as dansguardian?

Re: [coLinux-users] CoLinux/Dansguardian on Windows machine

[<Use...@zo...>]

dis...@bo...(dispose)  11.03.08 23:10

>Hi,
>For my children I have a dual boot machine with Linux and WIndowsXP.

>In Linux I use Dansguardian to keep the bad world out, 

Solving social problems with techinical matters never worked!
Educate your children, give them "media competence", surf together with them, 
but don't restrict them as dummies!
But ok, that's "your" children...


>but I've no such a reliable (and affordable) content filter for windows. 

>Would the following scenario be feasable:
>1. Setup colinux on XP
>2. Install DansGuardian/Squid on colinux
>3. Tell the windows browser (IE) to use colinux as proxy
>4. Let CoLinux connect to the web, and
>5. Let DansGuardian check the content.

>Important questions are further:
>1. Can CoLinux be started at XP-boot automatically

run it as a "service"

>2. How much will this approach affect the performance of the PC

If you have lot of memory you would not notice it while surfing.

>One further, a bit off topic, question:
>The PC has a AMD 2x 64 5000+ on board. 
>My winXP will only use one of the cores (I think). 

XP would use "SMP" with upto 2 CPU(core)s.
Have a look in taskmanager: You should see both cores there.
If not, you run the wrong kernel. Test MS-KB for more info.

You can inhibit that in boot.ini, see MS-KB.

>Would it be possible to make CoLinux use the other core?


>Thanks for your thoughts.


Have a look on "IPcop".
It should work under colinux too and there was "once" (1 year or
more ago) in heise c't an article howto to integrate IPcop into a 
VM in Windows.
I am quite sure that would be the easiest way to avoid reinventing 
the wheel ;-)

Rainer

Re: [coLinux-users] console-nt

[Henry N. <Hen...@Ar...>]

Paolo Minazzi wrote:
> hi colinux developers and users.
> 
> I have two things to communicate (a post and a question)
> 
> 1)
> with the 0.7.1 if I use
> stty rows n , where n>25 , Windows crash
> This is not an important problem, but it is an error.
> I tried to understand where is the problem (into the source) but I wan
> unabile to do it. I'm sorry :=(

Thanks. Live console resizing is not supported for colinux.

The console buffer is allocated on the windows side on starting coLinux
and can not change at live time. The patch will fix it.

-- 
Henry N.

Re: [coLinux-users] console-nt

[Paolo M. <pao...@gm...>]

On Tue, Mar 11, 2008 at 9:56 PM, Henry Nestler <Hen...@ar...> wrote:
> Hello,
>
>  is console size (NT and FLTK) an interesting option for coLinux?
>  Should we give it as runtime parameter? Instead the source code change
>  and recompiling.

I think that a #define with SIZE_X and SIZE_Y can be good.
Configuring the size at sartup can be good (an option vga=100x40, for
example) , but I think require some works .....
For advanced use there is putty.
Bye,
Paolo

Re: [coLinux-users] DamnSmallLinux on colinux

[Henry N. <Hen...@Ar...>]

insomniux wrote:
> I'm trying to setup DSL on colinux has someone done that before? I'd
> like to share experiences.
> 
> I've tried to follow the howto on installation of new distro;s, but I
> run into errors (see below).
> 
> My configuration is:
> WinXP SP2
> Colinx 0.7.2
> dsl-4.2.5-initrd.iso
> 
> I've copied the minirt24.gz from the iso file
> 
> 
> =====Colinux-Daemon-command=========
> C:\progra~1\colinux\colinux-daemon @dsl01.conf cofs0=d:\mike\colinux\dsl
> root=/dev/ram0 -t nt 
> 
> =====dsl01-setup.conf===============
> kernel=c:\PROGRA~1\coLinux\vmlinux
> cobd0="D:\Mike\colinux\dsl\dsl-4.2.5-initrd.iso"
> hda1="D:\Mike\colinux\dsl\dsl.fs"
> hda2="D:\Mike\colinux\dsl\swap_512Mb"
> root=/dev/hda1
> ro
> initrd="D:\Mike\colinux\dsl\minirt24.gz"
> mem=256
> eth0=pcap-bridge,"Colinux-TAP-host" 
> eth1=tuntap,"Colinux-TAP-guest"
> 
> =====Windows-console-output=========
> D:\Mike\colinux\dsl>C:\progra~1\colinux\colinux-daemon @dsl01.conf
> cofs0=d:\mik
> \colinux\dsl root=/dev/ram0 -t nt #ramdisk_size=40000
> Cooperative Linux Daemon, 0.7.2
> Compiled on Jan 18 2008 21:47:09
> 
> ...
> Kernel command line: root=/dev/hda1 ro root=/dev/ram0
> #ramdisk_size=40000

Typed wrong or you wand not use "ramdisk_size=..." ?

> ...
> RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize

4096 blocks a 1024 was set from coLinux kernel config:
CONFIG_BLK_DEV_RAM_SIZE=4096
CONFIG_BLK_DEV_RAM_BLOCKSIZE=1024

> ...
> RAMDISK: Compressed image found at block 0
> RAMDISK: incomplete write (-28 != 32768) 4194304

The initial error: Write to ramdisk failed.

> VFS: Mounted root (ext2 filesystem) readonly.
> Freeing unused kernel memory: 128k freed
> attempt to access beyond end of device
> ram0: rw=0, want=32776, limit=8192

Your initrd size is to small. You have try to increase the ram disk
size? That should help.
What should do the '#' there?
"#ramdisk_size=40000" --> "ramdisk_size=40000"

Perhaps more ideas for you on Colonist?
I found in old user mail (2005-02-06)
http://sourceforge.net/mailarchive/message.php?msg_id=cu4pt5%24872%241%40sea.gmane.org

-- 
Henry N.

Re: [coLinux-users] Block devices disappear

[H. N. <hen...@ar...>]

Hello,

this can be the same problem, I found with partitions under my SuSE.
You can see the error, if you would start colinux-debug-daemon before starts colinux (see in debugging.txt).

Please try coLinux version 0.7.3 from snapshots. There have fixed a fault while getting partition sizes. An opened device blocked it self from getting size information. The error was mostly for partitions, and a raice condition how the Linux kernel opens and closed the block device.

http://colinux.svn.sourceforge.net/viewvc/colinux?view=rev&revision=904

-- 
Henry

Tyler Larson wrote:
> Everything worked great till I started screwing around with adding block 
> devices getting ready to transfer everything to a disk partitions; now 
> block devices--seemingly at random--aren't available on Linux boot. I 
> haven't been able to figure out any consistent pattern to determine 
> which block devices will be visible within Linux and which will be 
> missing, but once a device disappears, it rarely reappears on subsequent 
> boots. I don't believe that the swap image ever disappeared before I 
> tried including disk partitions, but now it isn't even visible when I 
> revert back to my original configuration.
> 
> By "not visible" I mean that the device doesn't show up in 
> /proc/partitions. I've tried uninstalling and re-installing coLinux 
> itself and the linux.sys driver (rebooting between uninstall and 
> reinstall). I've tried booting with init=/bin/bash to make sure that the 
> initscripts weren't messing anything up.
> 
> Host OS is Windows Vista, running on a Core2 Duo laptop. I think all the 
> other information you need can be seen in the output below. Each of the 
> "missing" block devices has successfully been mounted in coLinux at 
> least once, but devices do "disappear" even if they've never been used 
> before--i.e. if I create a new empty file then try to include it as a 
> block device, it often will not be visible in Linux.
> 
> Below is an example of what I'm talking about. Notice that it 
> successfully maps the devices before the kernel boots, but only cobd0 
> appears in /proc/partitions. Let me know if there's any further 
> debugging information I can supply.
> 
> Cheers
> -Tyler
> 
> -----------------------------------------------------------
> C:\Program Files\coLinux>colinux-daemon.exe kernel=vmlinux 
> cobd0=fedora7\Fedora-
> 7.img cobd1=fedora7\swap.img cobd2="\Device\Harddisk0\Partition3" 
> cobd3="\Device
> \Harddisk0\Partition4" mem=512 root=/dev/cobd0 ro -v 3 -t nt
> Cooperative Linux Daemon, 0.7.2
> Compiled on Jan 18 2008 21:47:09
> 
> using 'vmlinux' as kernel image
> configuring 512 MB of virtual RAM
> mapping cobd0 to \??\C:\Program Files\coLinux\fedora7\Fedora-7.img
> mapping cobd1 to \??\C:\Program Files\coLinux\fedora7\swap.img
> mapping cobd2 to \Device\Harddisk0\Partition3
> mapping cobd3 to \Device\Harddisk0\Partition4
> kernel boot parameters: 'root=/dev/cobd0 ro'
> PID: 5096
> colinux: launching console
> colinux: booting
> Linux version 2.6.22-co-0.7.2 (hn@hn-lt) (gcc version 4.1.2) #1 PREEMPT 
> Fri Jan
> 18 21:44:04 UTC 2008
> 512MB LOWMEM available.
> Entering add_active_range(0, 0, 131072) 0 entries of 256 used
> Zone PFN ranges:
>   DMA             0 ->        0
>   Normal          0 ->   131072
> 
> ...
> -----------------------------------------------------------
> 
> [root@colinux ~]$ cat /proc/partitions
> major minor  #blocks  name
> 
>  117     0    4095968 cobd0
> [root@colinux ~]$
> 
> -----------------------------------------------------------


Jetzt komfortabel bei Arcor-Digital TV einsteigen: Mehr Happy Ends, mehr Herzschmerz, mehr Fernsehen! Erleben Sie 50 digitale TV Programme und optional 60 Pay TV Sender, einen elektronischen Programmführer mit Movie Star Bewertungen von TV Movie. Außerdem, aktuelle Filmhits und prickelnde Erotik in der Arcor-Videothek. Infos unter www.arcor.de/tv

[coLinux-users] Block devices disappear

[Tyler L. <ty...@tl...>]

Everything worked great till I started screwing around with adding block 
devices getting ready to transfer everything to a disk partitions; now 
block devices--seemingly at random--aren't available on Linux boot. I 
haven't been able to figure out any consistent pattern to determine 
which block devices will be visible within Linux and which will be 
missing, but once a device disappears, it rarely reappears on subsequent 
boots. I don't believe that the swap image ever disappeared before I 
tried including disk partitions, but now it isn't even visible when I 
revert back to my original configuration.

By "not visible" I mean that the device doesn't show up in 
/proc/partitions. I've tried uninstalling and re-installing coLinux 
itself and the linux.sys driver (rebooting between uninstall and 
reinstall). I've tried booting with init=/bin/bash to make sure that the 
initscripts weren't messing anything up.

Host OS is Windows Vista, running on a Core2 Duo laptop. I think all the 
other information you need can be seen in the output below. Each of the 
"missing" block devices has successfully been mounted in coLinux at 
least once, but devices do "disappear" even if they've never been used 
before--i.e. if I create a new empty file then try to include it as a 
block device, it often will not be visible in Linux.

Below is an example of what I'm talking about. Notice that it 
successfully maps the devices before the kernel boots, but only cobd0 
appears in /proc/partitions. Let me know if there's any further 
debugging information I can supply.

Cheers
-Tyler

-----------------------------------------------------------
C:\Program Files\coLinux>colinux-daemon.exe kernel=vmlinux 
cobd0=fedora7\Fedora-
7.img cobd1=fedora7\swap.img cobd2="\Device\Harddisk0\Partition3" 
cobd3="\Device
\Harddisk0\Partition4" mem=512 root=/dev/cobd0 ro -v 3 -t nt
Cooperative Linux Daemon, 0.7.2
Compiled on Jan 18 2008 21:47:09

using 'vmlinux' as kernel image
configuring 512 MB of virtual RAM
mapping cobd0 to \??\C:\Program Files\coLinux\fedora7\Fedora-7.img
mapping cobd1 to \??\C:\Program Files\coLinux\fedora7\swap.img
mapping cobd2 to \Device\Harddisk0\Partition3
mapping cobd3 to \Device\Harddisk0\Partition4
kernel boot parameters: 'root=/dev/cobd0 ro'
PID: 5096
colinux: launching console
colinux: booting
Linux version 2.6.22-co-0.7.2 (hn@hn-lt) (gcc version 4.1.2) #1 PREEMPT 
Fri Jan
18 21:44:04 UTC 2008
512MB LOWMEM available.
Entering add_active_range(0, 0, 131072) 0 entries of 256 used
Zone PFN ranges:
  DMA             0 ->        0
  Normal          0 ->   131072

...
-----------------------------------------------------------

[root@colinux ~]$ cat /proc/partitions
major minor  #blocks  name

 117     0    4095968 cobd0
[root@colinux ~]$

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

[coLinux-users] DamnSmallLinux on colinux

[insomniux <dis...@bo...>]

Hi,
I'm trying to setup DSL on colinux has someone done that before? I'd
like to share experiences.

I've tried to follow the howto on installation of new distro;s, but I
run into errors (see below).

My configuration is:
WinXP SP2
Colinx 0.7.2
dsl-4.2.5-initrd.iso

I've copied the minirt24.gz from the iso file


=====Colinux-Daemon-command=========
C:\progra~1\colinux\colinux-daemon @dsl01.conf cofs0=d:\mike\colinux\dsl
root=/dev/ram0 -t nt 

=====dsl01-setup.conf===============
kernel=c:\PROGRA~1\coLinux\vmlinux
cobd0="D:\Mike\colinux\dsl\dsl-4.2.5-initrd.iso"
hda1="D:\Mike\colinux\dsl\dsl.fs"
hda2="D:\Mike\colinux\dsl\swap_512Mb"
root=/dev/hda1
ro
initrd="D:\Mike\colinux\dsl\minirt24.gz"
mem=256
eth0=pcap-bridge,"Colinux-TAP-host" 
eth1=tuntap,"Colinux-TAP-guest"

=====Windows-console-output=========
D:\Mike\colinux\dsl>C:\progra~1\colinux\colinux-daemon @dsl01.conf
cofs0=d:\mik
\colinux\dsl root=/dev/ram0 -t nt #ramdisk_size=40000
Cooperative Linux Daemon, 0.7.2
Compiled on Jan 18 2008 21:47:09

PID: 888
colinux: booting
Linux version 2.6.22-co-0.7.2 (hn@hn-lt) (gcc version 4.1.2) #1 PREEMPT
Fri Jan
18 21:44:04 UTC 2008
256MB LOWMEM available.
initrd enabled: start: 0xccf35000  size: 0x030ca520
Entering add_active_range(0, 0, 65536) 0 entries of 256 used
Zone PFN ranges:
  DMA             0 ->        0
  Normal          0 ->    65536
early_node_map[1] active PFN ranges
    0:        0 ->    65536
On node 0 totalpages: 65536
  DMA zone: 0 pages used for memmap
  Normal zone: 512 pages used for memmap
  Normal zone: 65024 pages, LIFO batch:15
Built 1 zonelists.  Total pages: 65024
Kernel command line: root=/dev/hda1 ro root=/dev/ram0
#ramdisk_size=40000
Enabling fast FPU save and restore... done.
Enabling unmasked SIMD FPU exception support... done.
Initializing CPU#0
Setting proxy interrupt vectors
PID hash table entries: 1024 (order: 10, 4096 bytes)
Console: colour CoCON 80x25
Dentry cache hash table entries: 32768 (order: 5, 131072 bytes)
Inode-cache hash table entries: 16384 (order: 4, 65536 bytes)
Memory: 205952k/262144k available (1628k kernel code, 0k reserved, 591k
data, 1
8k init, 0k highmem)
virtual kernel memory layout:
    fixmap  : 0xffffc000 - 0xfffff000   (  12 kB)
    vmalloc : 0xd0800000 - 0xffffa000   ( 759 MB)
    lowmem  : 0xc0000000 - 0xd0000000   ( 256 MB)
      .init : 0xc032e000 - 0xc034e000   ( 128 kB)
      .data : 0xc02971f8 - 0xc032afe4   ( 591 kB)
      .text : 0xc0100000 - 0xc02971f8   (1628 kB)
Calibrating delay loop... console: Coopeartive Linux console started
colinux-net-daemon: searching TAP device named "Colinux-TAP-guest"
colinux-net-daemon: found TAP device named "Colinux-TAP-host"
colinux-net-daemon: found TAP device named "Colinux-TAP-guest"
colinux-net-daemon: opening TAP: "Colinux-TAP-guest"
colinux-net-daemon: TAP driver version 8.4
colinux-net-daemon: enabling TAP...
conet-bridged-daemon: Looking for interface "Colinux-TAP-host"
conet-bridged-daemon: adapter Adapter for generic dialup and VPN capture
doesn'
 have a connection
conet-bridged-daemon: checking connection: Colinux-TAP-guest
conet-bridged-daemon: checking connection: Real
conet-bridged-daemon: checking connection: Netwerkbrug (Netwerkbrug) 9
conet-bridged-daemon: checking connection: Colinux-TAP-host
conet-bridged-daemon: listening on: TAP-Win32 Adapter V8 (coLinux)...
conet-bridged-daemon: listening for: (ether dst 00:ff:0e:b0:00:00) or
(ether br
adcast or multicast) or (ip broadcast or multicast)
1071.51 BogoMIPS (lpj=5357568)
Mount-cache hash table entries: 512
CPU: After generic identify, caps: bfebfbff 00000000 00000000 00000000
00004400
00000000 00000000
CPU: Trace cache: 12K uops, L1 D cache: 8K
CPU: L2 cache: 512K
CPU: After all inits, caps: bfebf3ff 00000000 00000000 0000b080 00004400
000000
0 00000000
Compat vDSO mapped to ffffe000.
CPU: Intel(R) Xeon(TM) CPU 2.80GHz stepping 07
Checking 'hlt' instruction... OK.
NET: Registered protocol family 16
NET: Registered protocol family 2
IP route cache hash table entries: 2048 (order: 1, 8192 bytes)
TCP established hash table entries: 8192 (order: 4, 65536 bytes)
TCP bind hash table entries: 8192 (order: 3, 32768 bytes)
TCP: Hash tables configured (established 8192 bind 8192)
TCP reno registered
checking if image is initramfs...it isn't (no cpio magic); looks like an
initrd
Freeing initrd memory: 49961k freed
cofuse init 0.1 (API version 2.2)
io scheduler noop registered
io scheduler anticipatory registered (default)
io scheduler deadline registered
io scheduler cfq registered
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
cobd: loaded (max 32 devices)
alias for cobd1 is hda1
cobd alias cobd1 -> hda1 created
alias for cobd2 is hda2
cobd alias cobd2 -> hda2 created
loop: module loaded
conet: loaded (max 16 devices)
conet0: initialized
conet1: initialized
serio: cokbd at irq 1
mice: PS/2 mouse device common for all mice
input: AT Translated Set 2 keyboard as /class/input/input0
TCP cubic registered
NET: Registered protocol family 1
NET: Registered protocol family 17
Using IPI Shortcut mode
RAMDISK: Compressed image found at block 0
RAMDISK: incomplete write (-28 != 32768) 4194304
VFS: Mounted root (ext2 filesystem) readonly.
Freeing unused kernel memory: 128k freed
attempt to access beyond end of device
ram0: rw=0, want=32776, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block
- inode
3937, block=16387
Warning: unable to open an initial console.
attempt to access beyond end of device
ram0: rw=0, want=65544, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block
- inode
7875, block=32771
attempt to access beyond end of device
ram0: rw=0, want=65544, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block
- inode
7873, block=32771
attempt to access beyond end of device
ram0: rw=0, want=16396, limit=8192
EXT2-fs error (device ram0): ext2_get_inode: unable to read inode block
- inode
1969, block=8197
Kernel panic - not syncing: No init found.  Try passing init= option to
kernel.
colinux: Linux VM terminated
colinux: terminated with code 3 - abnormal exit, aborting
console: Monitor888: Detached

D:\Mike\colinux\dsl>
=======================================

Thanks
Mike

Re: [coLinux-users] Virtual Drive issues

[Henry N. <Hen...@Ar...>]

Andy Olsen wrote:
> I am using andLinux which uses colinux. Every time I boot up my
> computer, the colinux-daemon rewrites my virtual drive, base.drv. I
> type in poweroff before I reboot/restart my computer and this doesn't
> seem to help. I have also noticed that this virtual drive rewrite
> occurs randomly even though this virtual drive has rewritten since I
> have powered on my computer. Any suggestions how to resolve this
> issue? This issue slows down my computer while the colinux-daemon
> rewrites 4 GB of data to my hard drive.

Check the settings in /etc/fstab, that the names of your Root device is ok.

Verify the filesystem parameters with "tune2fs -l /dev/cobdX"  (with is
X your Root filesystem). Locate the Last check time and check interval,
and the mount count with the max mount count. Both options you can
increase or disable with -i and -c.

fsck depens on days between last check. Check the time/date of your
Linux and Windows. A large or negative timediff triggers the fsck.

Remove the "initrd=initrd.gz", if you have it.
An other idea is to add "ro" to your bootparameters.

-- 
Henry N.

[coLinux-users] Virtual Drive issues

[Andy O. <cod...@gm...>]

I am using andLinux which uses colinux. Every time I boot up my
computer, the colinux-daemon rewrites my virtual drive, base.drv. I
type in poweroff before I reboot/restart my computer and this doesn't
seem to help. I have also noticed that this virtual drive rewrite
occurs randomly even though this virtual drive has rewritten since I
have powered on my computer. Any suggestions how to resolve this
issue? This issue slows down my computer while the colinux-daemon
rewrites 4 GB of data to my hard drive.

Thanks.

Re: [coLinux-users] newbie trying to get started with colinux

[Bill C. R. <do...@gm...>]

Alex,

I can't speak to the problem of being blind, but when I started using
coLinux, it was because I had severe hand pains and could not type.  
The best quality voice software is only available for Windows.   So in
order to continue to use Linux, I had to be creative.   In the end, I
found using programs like emacs running under coLinux and redirected to
Windows worked reasonably well with voice recognition software.  
However, by the time I started to figure that out my hands were already
recovering.   After treatment and switching to Dvorak, I have been able
to type normally, so I never finished the process of learning emacs speak.

One thing to consider, is you don't really need the Linux system to
speak.  If you connect to the Linux system via emacs under windows, then
you will only need to have the Windows version of emacs do screen reading.

If you need help setting up your system, contact me off line.   I have
never setup a system this way, but I am willing to help you figure it
out.   If we can get something working, we can then post the image to
help anyone else who needs it as well.

Bill
 
alex wallis wrote:
> Hi. Thanks for replying to my message.
> I've looked at the notes, but i'm not sure how do i create the batch file it 
> says i need to make? and where should this file be put?
> also what does this step mean?
> "5. Setup your keyboard with "dpkg-reconfigure console-data"
> I no this is a long shot, but is there anyone on this list who is blind? as 
> i'm blind and and because of this My first need after trying to get 
> everything set up will be to get some kind of screen reader working such as 
> linux-speakup or orca.
> I've heard that getting a linux screen reader working in this kind of 
> situation can be hard so just wondering has anyone had experience of this?
> ----- Original Message ----- 
> From: "Henry Nestler" <Hen...@Ar...>
> To: "alex wallis" <alexwallis646@google    mail.com>
> Cc: <col...@li...>
> Sent: Monday, February 25, 2008 5:31 PM
> Subject: Re: [coLinux-users] newbie trying to get started with colinux
>
>
>   
>> alex wallis wrote:
>>     
>>> Hi.
>>> `    My name is Alex. I recently found out about colinux on another 
>>> mailing list.
>>> Am I right in thinking that colinux takes up a lot less resources than 
>>> say
>>> running vmware?
>>> I'm writing because I have installed colinux, I have downloaded one of 
>>> the
>>> distribution options given in the installer, but now i'm not really sure
>>> where to start with getting everything working.
>>> I have successfully extracted the image that the installer downloaded.
>>> I was also wondering in the ubuntu image I have downloaded is that just a
>>> basic file system? or does it include packages such as orca.
>>> Sorry for all the questions.
>>> I did try to check the wiki but the site seems to be down at the moment.
>>>       
>> Yust in the moment, the wiki is online.
>>
>> The image files in installer are perhaps a liddle outdated. So, it is
>> very goot, you have downloaded a more recent image from sf.net.
>>
>> If you have downloaded and unpacked the Ubuntu image, than should follow
>> the notes from
>> http:/
>>     
> /sourceforge.net/project/shownotes.php?release_id=544248&group_id=98788
>   
>> That contains all you need to run the Ubuntu. In the end you have a
>> minimal Linux system, that is usable to install more packages in same
>> way, as you installed Ubuntu nativly. The packages are not in the
>> filesystem. Ubuntu use an online repository.
>>
>> The difference between VMware and coLinux is mainly, that coLinux does
>> no needs to emulate any hardware, and coLinux is very fair with memory
>> usage. You can run coLinux with 128MB for simple text consoles. And if
>> Linux give free some memory, than coLinux give that free to toe host.
>>
>> Here is one of my fast booted coLinux
>> http://www.henrynestler.com/colinux/screenshoots/bootchart-async-38sec.png,
>> that is a SuSE9.0 standard installation on a partition, used as dual
>> boot and coLinux. VMware would not good run on this machine (256MB total
>> RAM). More faster boots are Debian 3.0 or latest ArchLinux.
>>
>> -- 
>> Henry N. 
>>     
>
>
> -------------------------------------------------------------------------
> This SF.net email is sponsored by: Microsoft
> Defy all challenges. Microsoft(R) Visual Studio 2008.
> http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
> _______________________________________________
> coLinux-users mailing list
> coL...@li...
> https://lists.sourceforge.net/lists/listinfo/colinux-users
>   

Re: [coLinux-users] CoLinux/Dansguardian on Windows machine

[Holger K. <hol...@gm...>]

dispose schrieb:
> reliable (and affordable) content filter for windows. Would the 
> following scenario be feasable:
> 1. Setup colinux on XP
> 2. Install DansGuardian/Squid on colinux
> 3. Tell the windows browser (IE) to use colinux as proxy
> 4. Let CoLinux connect to the web, and
> 5. Let DansGuardian check the content.

Yes, that will work.


> Important questions are further:
> 1. Can CoLinux be started at XP-boot automatically

Yes.

> 2. How much will this approach affect the performance of the PC

Not much. You loose 128 MB of memory. And if you have a stripped down linux it will not use the cpu if squid/dansguardian has nothing to do.


> 
> One further, a bit off topic, question:
> The PC has a AMD 2x 64 5000+ on board. My winXP will only use one of the 
> cores (I think). Would it be possible to make CoLinux use the other core?

XP uses both cores, check with task manager if you have 2 cpuload graphs. If not something with your install went wrong. colinux can only use what windows recognizes.

[coLinux-users] CoLinux/Dansguardian on Windows machine

[dispose <dis...@bo...>]

Hi,
For my children I have a dual boot machine with Linux and WIndowsXP. In
Linux I use Dansguardian to keep the bad world out, but I've no such a
reliable (and affordable) content filter for windows. Would the
following scenario be feasable:
1. Setup colinux on XP
2. Install DansGuardian/Squid on colinux
3. Tell the windows browser (IE) to use colinux as proxy
4. Let CoLinux connect to the web, and
5. Let DansGuardian check the content.

Important questions are further:
1. Can CoLinux be started at XP-boot automatically
2. How much will this approach affect the performance of the PC

One further, a bit off topic, question: 
The PC has a AMD 2x 64 5000+ on board. My winXP will only use one of the
cores (I think). Would it be possible to make CoLinux use the other
core? 

Thanks for your thoughts.
Insomniux

Re: [coLinux-users] console-nt

[Henry N. <Hen...@Ar...>]

Hello,

is console size (NT and FLTK) an interesting option for coLinux?
Should we give it as runtime parameter? Instead the source code change
and recompiling.

Henry Nestler wrote:
> Hello Paolo,
> 
> think, you have forgotten two lines with 79 and 24, the "values-1".
> Have changed console size to 40x120. Hope, it is not to big now. ;-)
> 
> Here is the complete patch for coLinux 0.8.0:
> http://www.henrynestler.com/colinux/patches/devel/console-size-40x120.patch
> Should also be usable for 0.7.2/0.7.3, but not tested.
> 
> and a ready to use build with it:
> http://www.henrynestler.com/colinux/testing/devel-0.8.0/20080310-con-40x120/
> 
> and a screenshot:
> http://www.henrynestler.com/colinux/screenshoots/Console-size-40x120.png
> 
> Paolo Minazzi wrote:
>> hi colinux developers and users.
>>
>> I have two things to communicate (a post and a question)
>>
>> 1)
>> with the 0.7.1 if I use
>> stty rows n , where n>25 , Windows crash
>> This is not an important problem, but it is an error.
>> I tried to understand where is the problem (into the source) but I wan
>> unabile to do it. I'm sorry :=(
>>
>> 2)
>> I have tried to adjust the console-nt to have a larger console, for
>> example 80x40.
>> I have modified some lines in the code :
>>
>> src/colinux/kernel/monitor.c:915:
>>     rc = co_console_create(80, 40 /* 25 */ , 0, &cmon->console);  // BY PAOLO
>>
>> src/colinux/os/linux/kernel/module/.module.kernel_build/monitor.c:915:
>>     rc = co_console_create(80, 40 /* 25 */ , 0, &cmon->console);  // BY PAOLO
>>
>> src/colinux/os/current/user/console-nt/widget.cpp:125
>>     r.bottom = fs.Y * 40; // 25; BY PAOLO
>>
>> src/colinux/os/current/user/console-nt/widget.cpp:141
>>     size.Y = 40 ; // 25 ; PAOLO
>>
>> ~/build/linux-2.6.22-source/drivers/video/console:63
>>     c->vc_rows = 40; // 25; BY PAOLO
>>
>> recompile all.
>> Colinux on Win32 starts ok, but afterm some messages the system crash
>> and my PC reboots.
>>
>> I have searched something like "4000" (=80x25) or "8000" (=80*25*2,
>> for char and attribute) but nothing.
>>
>> I know that it is possbile use putty, but the console sometimes is more useful.
>> If someone have ideas, please submit it.
>>
>> Regards,
>> Paolo Minazzi
>>

-- 
Henry N.