[uml-user] generic ebtables script ?

[Sunny Dubey <sunny-ml@op...>]

hi guys

I was wondering if there was some sample/generic ebtables script ?  Something 
someone with UMLs would be interested in using.  I'd like to have some 
direction and something to learn from :^)

Or should I skip ebtables altogether and use the brnf patch ?

Thanks for any advice

Sunny Dubey

Re: [uml-user] Slow NFS between UML and host

[David Cannings <david@fu...>]

Goetz Bock wrote:
> On Thu, Oct 28 '04 at 19:54, David Cannings wrote:
>> Mounting the share is fast (portmap is definitely running) and simple
>> commands such as ls are instantaneous too.  However, running 'emerge
>> search <anything>' or 'emerge -avuD world' takes much longer than it
>> does anywhere else on the network.
> As this has not been fixed, it affexts every use of UDP, and NFS is one.

As I mentioned in my first email I have tried the option 'tcp' to mount.

> You can work arround the bug by using rsize/wsize=1500 to keep the UDP
> package small.

I have tried forcing the size to <=1500 but I see no difference.  The
speed is still much slower than other machines on the network.

David

Re: [uml-user] Slow NFS between UML and host

[Goetz Bock <bock+uml@bl...>]

On Thu, Oct 28 '04 at 19:54, David Cannings wrote:
> Mounting the share is fast (portmap is definitely running) and simple
> commands such as ls are instantaneous too.  However, running 'emerge
> search <anything>' or 'emerge -avuD world' takes much longer than it
> does anywhere else on the network.

if you search the arcive you'll see that there was a problem with big
UDP packages mentioned 2-3 weeks ago.

As this has not been fixed, it affexts every use of UDP, and NFS is one.

You can work arround the bug by using rsize/wsize=1500 to keep the UDP
package small.

As long as it does not get fragmented, everything is fine.
-- 
/"\ Goetz Bock at blacknet dot de  --  secure mobile Linux everNETting
\ /       (c) 2004 Creative Commons, Attribution-ShareAlike 2.0 de
 X   [ 1. Use descriptive subjects - 2. Edit a reply for brevity -  ]
/ \  [ 3. Reply to the list - 4. Read the archive *before* you post ]

Re: [uml-user] how to use a bochs .img as root_fs ?

[Andreas Wohlfeld <agw@su...>]

On Sun, Oct 31, 2004 at 10:14:27AM +0100, Andrea Tasso wrote:
> I used bochs to make an image file, say woody.img, with a running debian
> woody on it. I was able to use the that filesystem for uml also. I just

If bochs img is not a paritition file but a complete harddisk file, you
would have to use the partitions within uml also.
Normally you use your harddisk files in uml as a super floppy, not
partitioning it, but you can do so.
I think you have to create some ubd devices which correspond to the
partitions and then mount these in fstab.

Never tried it, dunno the devices.

-- 
Facts do not cease to exist because they are ignored.
  - Aldous Huxley

[uml-user] UML performance and safety

[The NeverGone <never@de...>]

Hy...!

What can i do in favour of this situation, when i would like to use UML 
for network's goal with higher level of achievement and safety. I know the 
patch called "-skas3", but i don't know that i could use the "-devanon"'s 
patch in this case, or rather how the '-djc' and "-sysemu" patchs could 
help for the system of host; and the "-um" patch for the UML guest-system. 
Could you give me any helpful advice how to raise the power? Could i suit 
these patchs with the security system of GrSecurity (and Pax)...?
The host-system Debian GNU/Linux (Woody - stable) with Linux-kernel 
2.4.27, it would be a Linux system which possesses a Linux-kernel 2.6.9 on 
the UML system.


Thanks:
Kurucz "The NeverGone" Istvan :)

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
  Ici-pici blogocskam: http://nevergone.freeblog.hu/
  LinuxCounter: #355477
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

[uml-user] how to use a bochs .img as root_fs ?

[Andrea Tasso <andrea@ta...>]

hi all,
I used bochs to make an image file, say woody.img, with a running debian
woody on it. I was able to use the that filesystem for uml also. I just
loop mounted woody.img and an empy new root_fs, and copied everything
from woody.img to root_fs. Then, with some small changes, root_fs is a
very good fs for uml. To mount woody.img I had to use a trick:
mount -o loop offset=32256 -t ext2 woody.img woody-mnt/
You can easily guess I cannot (immediately) use woody.img as the root fs
for uml, even if I do some small changes, because at boot uml complains
it cannot find the superblock.
What I know is that woody.img can be mounted with the offset=32256
option, but if I try to feed it to e2fsck, eg
e2fsck -b 32256 woody.img
I get the same "superblock not found" error, and also with a script I
passed to e2fsck a superblock location from 1 to 50000 but I had no
success.
I also reviewed kernel command line options, but I got no clue.
So how to use a bochs .img as a root_fs without having to copy the
filesystem to another image file ?
thanks

-- 
Andrea Tasso <andrea@...> (http://andrea.tasso.info)
GnuPG key: search for 'andrea@...' on
http://www.cam.ac.uk.pgp.net/pgpnet/wwwkeys.html
Key fingerprint: 1DD5 98C8 289A C1DF 7CFF  B676 BDB6 6DC9 AD6A 745E

[uml-user] SBUML project?

[Lars Kellogg-Stedman <lars@od...>]

Howdy,

Are you folks familiar with the SBUML project?

  http://sbuml.sourceforge.net/

It doesn't appear on the "Related links" page at 
<http://user-mode-linux.sourceforge.net/links.html>;, but it looks like 
an interesting project -- they've added code to UML to allow 
"snapshotting" of the virtual machine state, so you can create an image 
of a running system that will "boot" back up exactly where it left off.

They have some really interesting ideas, and it looks as if development 
on this project was active as recently as July.

I found this project via a published paper here (long URL, so apologies 
of it wraps):

  
http://www.usenix.org/publications/library/proceedings/vm04/wips/potter.p
df

-- Lars

Re: [uml-user] uml as shared lib?

[Leendert Meyer <leen.meyer@ho...>]

On Saturday 30 October 2004 13:14, Leendert Meyer wrote:
> On Saturday 30 October 2004 12:24, Adil  Mujeeb, Noida wrote:
> > Hi,
> > I read that UML can be used as shared library.
>
> Can you give a pointer to where you read that?
>
> > how ? any practical example.

On Saturday 30 October 2004 13:45, you wrote:

> here is the link
>
> http://user-mode-linux.sourceforge.net/UserModeLinux-HOWTO-1.html
>
> section 1.2

Indeed:

3. Flexible code. Normally an OS kernel is just that... a kernel. 
   It talks to hardware or maybe some virtualised hardware. But 
   UML can be viewed in many other ways. It would be possible to 
   turn it into a shared library, for example, so that other 
   programs could link to it to take advantage of things that 
   Linux does very well. It can be started as a subshell of an 
   existing application. It can use stin/stdout like any other 
   program.

Cheers,

Leen

RE: [uml-user] uml as shared lib?

[Adil Mujeeb, Noida <adilm@hc...>]

here is the link

http://user-mode-linux.sourceforge.net/UserModeLinux-HOWTO-1.html

section 1.2

Thanks and Best Regards, 
Adil Mujeeb
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Contributing to the World by creating indispensable value ! 
System Software CoE @ HCLT-Noida 
HCL Technologies Ltd., 
A-11, Sector-16, Noida-201 301 
http://www.hcltechnologies.com
Ph. : +91-120-2510701/702 Ext : 3016
FAX : +91-120-2510713
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



-----Original Message-----
From: user-mode-linux-user-admin@...
[mailto:user-mode-linux-user-admin@...]On Behalf Of
Leendert Meyer
Sent: Saturday, October 30, 2004 4:45 PM
To: user-mode-linux-user@...
Subject: Re: [uml-user] uml as shared lib?


On Saturday 30 October 2004 12:24, Adil  Mujeeb, Noida wrote:
> Hi,
> I read that UML can be used as shared library.

Can you give a pointer to where you read that?

> how ? any practical example. 

Cheers,

Leen


-------------------------------------------------------
This SF.Net email is sponsored by:
Sybase ASE Linux Express Edition - download now for FREE
LinuxWorld Reader's Choice Award Winner for best database on Linux.
http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user@...
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user

Re: [uml-user] uml as shared lib?

[Leendert Meyer <leen.meyer@ho...>]

On Saturday 30 October 2004 12:24, Adil  Mujeeb, Noida wrote:
> Hi,
> I read that UML can be used as shared library.

Can you give a pointer to where you read that?

> how ? any practical example. 

Cheers,

Leen

[uml-user] uml as shared lib?

[Adil Mujeeb, Noida <adilm@hc...>]

Hi,
I read that UML can be used as shared library. how ? any practical example.

Thanks and Best Regards,
Adil Mujeeb

Re: [uml-user] How to get error log without x-term

[Stefan Nitz <uml@to...>]

Am Freitag, 29. Oktober 2004 22:44 schrieb Jimmy Pan:
> Hi,
>  i use xterm=none because i don't want the extra 2 x-terms come up at my
> server desktop...
> and one more thing, when i run UML and it came to
> .............
>
> ubdb: unknown partition table
>  ubdc: unknown partition table
The image has no partition table.
> Initializing stdio console driver
>
>
> it just STOP here...never proceed. do you know what's the problem??
>
> Jimmy
>
> >>                   \
> >>      xterm=none    \
> >
> >xterm=none will try to run the "none" executable in place of the "xterm"
> > one. So it won't work. In fact you don't see the consoles starting up.
> >
> >>      con0=fd:3 3>>/var/log/kernel.dump
> >>
> >>
> >>
> >>
> >>
> >>it runs like:
> >>
> >>
> >>
> >>
> >>
> >>root@...:/umls/test_v24# ./run_uml 105
> >>Set 'tap105' persistent and owned by uid 0
> >>SIOCDELRT: No such process
> >>SIOCADDRT: File exists
> >>Checking for the skas3 patch in the host...found
> >>Checking for /proc/mm...found
> >>
> >>
> >>
> >>
> >>
> >>Partition check:
> >> ubda: unknown partition table
> >> ubdb: unknown partition table
> >> ubdc: unknown partition table
> >>Initializing stdio console driver
> >
> >At this line, it has started using the driver handling all con= options,
> > so the subsequent output should be found in /var/log/kernel.dump.
> > However, either you run UML as root (which is very silly for security
> > reasons, but is what you're doing - I'd suggest you to stop that and run
> > it as normal user) or, if you run it as normal user, it won't be able to
> > access the file.
> >
> >Specify one which can be accessed by Uml running as user and run it as
> > user.
>
> -------------------------------------------------------
> This SF.Net email is sponsored by:
> Sybase ASE Linux Express Edition - download now for FREE
> LinuxWorld Reader's Choice Award Winner for best database on Linux.
> http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click
> _______________________________________________
> User-mode-linux-user mailing list
> User-mode-linux-user@...

-- 
Dear listeners,

Re: [uml-user] How to get error log without x-term

[Jimmy Pan <jpan@de...>]

Hi,
 i use xterm=none because i don't want the extra 2 x-terms come up at my 
server desktop...
and one more thing, when i run UML and it came to
.............

ubdb: unknown partition table
 ubdc: unknown partition table
Initializing stdio console driver


it just STOP here...never proceed. do you know what's the problem??

Jimmy

>>                   \
>>      xterm=none    \
>>    
>>
>xterm=none will try to run the "none" executable in place of the "xterm" one. 
>So it won't work. In fact you don't see the consoles starting up.
>  
>
>>      con0=fd:3 3>>/var/log/kernel.dump
>>    
>>
>
>  
>
>>it runs like:
>>    
>>
>
>  
>
>>root@...:/umls/test_v24# ./run_uml 105
>>Set 'tap105' persistent and owned by uid 0
>>SIOCDELRT: No such process
>>SIOCADDRT: File exists
>>Checking for the skas3 patch in the host...found
>>Checking for /proc/mm...found
>>    
>>
>
>  
>
>>Partition check:
>> ubda: unknown partition table
>> ubdb: unknown partition table
>> ubdc: unknown partition table
>>Initializing stdio console driver
>>    
>>
>At this line, it has started using the driver handling all con= options, so 
>the subsequent output should be found in /var/log/kernel.dump. However, 
>either you run UML as root (which is very silly for security reasons, but is 
>what you're doing - I'd suggest you to stop that and run it as normal user) 
>or, if you run it as normal user, it won't be able to access the file.
>
>Specify one which can be accessed by Uml running as user and run it as user.
>
>  
>

Re: [uml-user] Re: 2.6.9 guest - no make linux target

[Antoine Martin <antoine@na...>]

> > The 32-bit kernels now boot, but get stuck on starting init:
> > "./kernel32-2.6.9uml ubd0=./root_fs devfs=nomount root=/dev/ubda
> > mem=256M con0=fd:0,fd:1 init=/bin/bash
> > (...)
> > VFS: Mounted root (reiserfs filesystem) readonly.
> > init_new_context_skas - new_mm failed, errno = -2
> > init_new_context_skas - new_mm failed, errno = -2
> > init_new_context_skas - new_mm failed, errno = -2
> > Kernel panic - not syncing: No init found.  Try passing init= option to
> > kernel."
> > (call trace and register dump omitted)
> 
> Try adding mode=tt. The problem is that it tries to use /proc/mm even if that 
> is not supported by the host; mode=tt should force it to run in TT mode.
My mistake, the 32-bit kernels I had tried had skas but not TT, DOH!
I feel pretty stupid. (Thanks for your help)

> I wonder why that happened: have you got a /proc/mm file on your host?
AFAIK there is no such option for x86_64.

Thanks
Antoine

Re: [uml-user] tap device

[Anthony Brock <Anthony_Brock@ou...>]

>>> "Adil  Mujeeb, Noida" <adilm@...> 10/28/04 09:24PM >>>
> is there any way to delete the tap device while UML is running??

I would first try removing the device from the UML (with the listed =
command) and then attempt to delete the tap device.

Tony

Re: [uml-user] How to get error log without x-term

[Blaisorblade <blaisorblade_spam@ya...>]

On Friday 29 October 2004 19:49, Jimmy Pan wrote:
> Hi, when i tried to add the line to my run_uml script like the following:
> (check the last line)

> #!/bin/bash

> let gw=255-$1

> #echo "tap$1"
> tunctl -t tap$1 -u root
> ifconfig tap$1 192.168.0.$gw up
> route del -net 192.168.0.0 netmask 255.255.255.0
> route add -host 192.168.0.$1 dev tap$1
> echo 1 > /proc/sys/net/ipv4/ip_forward
> echo 1 > /proc/sys/net/ipv4/conf/tap$1/proxy_arp
> arp -Ds 192.168.0.$1 eth0 pub
> chmod 666 /dev/net/tun
>
> linux ubd0=cow_$1,fs_deb3C_2.ex2           \
>       ubd2=varfs_$1                          \
>       ubd1=swapfs_$1                         \
>       eth0=tuntap,tap$1                    \
>       mem=64M                             \
>       xterm=none    \
xterm=none will try to run the "none" executable in place of the "xterm" one. 
So it won't work. In fact you don't see the consoles starting up.
>       con0=fd:3 3>>/var/log/kernel.dump

> it runs like:

> root@...:/umls/test_v24# ./run_uml 105
> Set 'tap105' persistent and owned by uid 0
> SIOCDELRT: No such process
> SIOCADDRT: File exists
> Checking for the skas3 patch in the host...found
> Checking for /proc/mm...found

> Partition check:
>  ubda: unknown partition table
>  ubdb: unknown partition table
>  ubdc: unknown partition table
> Initializing stdio console driver
At this line, it has started using the driver handling all con= options, so 
the subsequent output should be found in /var/log/kernel.dump. However, 
either you run UML as root (which is very silly for security reasons, but is 
what you're doing - I'd suggest you to stop that and run it as normal user) 
or, if you run it as normal user, it won't be able to access the file.

Specify one which can be accessed by Uml running as user and run it as user.

-- 
Paolo Giarrusso, aka Blaisorblade
Linux registered user n. 292729

Re: [uml-user] How to get error log without x-term

[Jimmy Pan <jpan@de...>]

Hi, when i tried to add the line to my run_uml script like the following:
(check the last line)

#!/bin/bash

let gw=255-$1

#echo "tap$1"
tunctl -t tap$1 -u root
ifconfig tap$1 192.168.0.$gw up
route del -net 192.168.0.0 netmask 255.255.255.0
route add -host 192.168.0.$1 dev tap$1
echo 1 > /proc/sys/net/ipv4/ip_forward
echo 1 > /proc/sys/net/ipv4/conf/tap$1/proxy_arp
arp -Ds 192.168.0.$1 eth0 pub
chmod 666 /dev/net/tun

linux ubd0=cow_$1,fs_deb3C_2.ex2           \
      ubd2=varfs_$1                          \
      ubd1=swapfs_$1                         \
      eth0=tuntap,tap$1                    \
      mem=64M                             \
      xterm=none    \
      con0=fd:3 3>>/var/log/kernel.dump

it runs like:

root@...:/umls/test_v24# ./run_uml 105
Set 'tap105' persistent and owned by uid 0
SIOCDELRT: No such process
SIOCADDRT: File exists
Checking for the skas3 patch in the host...found
Checking for /proc/mm...found
..............................................
.................................................
...........................................
SGI XFS with ACLs, no debug enabled
SGI XFS Quota Management subsystem
aio_thread failed to initialize context, err = 38
2.6 AIO not supported on the host - reverting to 2.4 AIO
pty: 256 Unix98 ptys configured
RAMDISK driver initialized: 16 RAM disks of 4096K size 1024 blocksize
loop: loaded (max 8 devices)
Initializing Cryptographic API
Initializing software serial port version 1
mconsole (version 2) initialized on /root/.uml/UDKXrN/mconsole
Partition check:
 ubda: unknown partition table
 ubdb: unknown partition table
 ubdc: unknown partition table
Initializing stdio console driver


then , it was stuck here....
is there any problems in my script?


Jimmy



Blaisorblade wrote:

>On Wednesday 27 October 2004 23:58, Jimmy Pan wrote:
>  
>
>>Hi....if i don't use any console/x-term for UML, is there a way that i
>>can keep the error messages
>>in a log file? (My UMLs crashes very often....)
>>    
>>
>
>A stable setup can be achieved, even if it is not that easy (UML is very 
>dependant on a number of factors; as kernels often do not compile with some 
>gcc version, UML depends also onto glibc / binutils / NPTL support / host 
>kernel).
>  
>
>>Jimmy
>>    
>>
>If you mean messages from the main console, use:
>con0=fd:3 3>>LogFile
>
>Which will append the console output to LogFile. But the early output will not 
>be saved...
>
>Bye
>  
>

Re: [uml-user] interface device buzy problem

[Blaisorblade <blaisorblade_spam@ya...>]

On Friday 29 October 2004 14:06, Adil  Mujeeb, Noida wrote:
> Hi,
> one more observation
> in running UML, if i down the interface say eth0 as `ifconfig eth0 down`
> and later on want to up it (`ifconfig eth0 X.X.X.X up` ), it gives message
> that resource or device is buzy.

> SO is there no way to up the device once it is down????
It normally works. I guess that there can be some bugs, either in UML or in 
your setup, when you don't use the setuid uml_net, i.e. the simpler solution. 
I guess you can solve your problem that way. After doing this test, if you 
don't want to have uml_net setuid, 

> i tried to down it on host and then try to delete it thru tunctl -t tap0 ,
> it gives the same error message: TUNSETIFF Device or Resource is buzy.
> Only when i halt the UML then only it is deleted.
Well, even if you ifconfig the interface down, it is still used by UML. If you 
want to stop even that usage, try using uml_mconsole and its "remove " 
command, as in:

uml_mconsole <your UML's umid> remove eth0

> > Hi,
> > I have some question related to tuntap & bridge.
> > I have setup the bridge and started the UML. if i try to delete the tap
> > device it says " resource or device is buzy"
> > is there any way to delete the tap device while UML is running??
>
> You will probably need to bring it down on the host, too.
>
> # ifconfig mytap down
>
> David

-- 
Paolo Giarrusso, aka Blaisorblade
Linux registered user n. 292729

Re: [uml-user] Re: 2.6.9 guest - no make linux target

[Blaisorblade <blaisorblade_spam@ya...>]

On Friday 29 October 2004 04:02, Antoine Martin wrote:
> > >>UML is not the only app which breaks with NPTL. Why don't you try the
> > >> double libc road?
> > >
> > > Mostly, because I don't know how to properly set up glibc in this way
> > > - I'll have to take a look at the Gentoo ebuild, and see what I can
> > > come up with.  But, in 300-some odd packages I compile for my desktop
> > > systems, UML kernels is the only place I've seen this one come up.
> > > There's some minor problems with apps that link against db4 on a NPTL
> > > system, but these are easily worked around with a little LDFLAGS
> > > trickery.  Beyond that, just about everything I've seen compiles on a
> > > NPTL-only system.
> >
> > Just use one of the newer glibc ebuilds that are still ~x86. They build
> > two glibcs. You will recognize the ebuild by the new "nptlonly" use-flag.
> >
> >    ACCEPT_KEYWORDS="~x86" emerge glibc -vp
> >
> > should do the trick for you.
>
> Thanks it worked!
> I also needed to emerge "emul-linux-x86-baselibs", and enable "multilib"
> for glibc.
>
>
> Now, for the things that don't work quite right yet:
> The 32-bit kernels now boot, but get stuck on starting init:
> "./kernel32-2.6.9uml ubd0=./root_fs devfs=nomount root=/dev/ubda
> mem=256M con0=fd:0,fd:1 init=/bin/bash
> (...)
> VFS: Mounted root (reiserfs filesystem) readonly.
> init_new_context_skas - new_mm failed, errno = -2
> init_new_context_skas - new_mm failed, errno = -2
> init_new_context_skas - new_mm failed, errno = -2
> Kernel panic - not syncing: No init found.  Try passing init= option to
> kernel."
> (call trace and register dump omitted)

Try adding mode=tt. The problem is that it tries to use /proc/mm even if that 
is not supported by the host; mode=tt should force it to run in TT mode.

I wonder why that happened: have you got a /proc/mm file on your host?

> Note that the same root_fs works with the same kernels on a 32-bit host.
> Obviously the host kernel does not have skas (2.6.9 64-bit kernel)
> I tried all sorts of values for init= (bin/bash etc), with no effect.
> I tried various versions of the guest kernel (2.4.2x, 2.6.x).

> A 64-bit guest kernel based on 2.6.4+patches kinda works with the same
> settings and root_fs (ignoring unrelated process leak and other
> instability)...
The process leak is solvable with a 2.6.8.1 host. I'm going to put a fix for 
the normal UMLs to run on 2.6.9, but I guess that won't apply on a 64-bit UML 
kernel tree...
-- 
Paolo Giarrusso, aka Blaisorblade
Linux registered user n. 292729

[uml-user] interface device buzy problem

[Adil Mujeeb, Noida <adilm@hc...>]

Hi,
one more observation
in running UML, if i down the interface say eth0 as `ifconfig eth0 down` and
later on want to up it (`ifconfig eth0 X.X.X.X up` ), it gives message that
resource or device is buzy.

SO is there no way to up the device once it is down????

Thanks and Best Regards, 
Adil Mujeeb
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Contributing to the World by creating indispensable value ! 
System Software CoE @ HCLT-Noida 
HCL Technologies Ltd., 
A-11, Sector-16, Noida-201 301 
http://www.hcltechnologies.com
Ph. : +91-120-2510701/702 Ext : 3016
FAX : +91-120-2510713
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~



-----Original Message-----
From: Adil Mujeeb, Noida 
Sent: Friday, October 29, 2004 3:41 PM
To: user-mode-linux-user@...
Subject: RE: [uml-user] tap device


i tried to down it on host and then try to delete it thru tunctl -t tap0 ,
it gives the same error message: TUNSETIFF Device or Resource is buzy.
Only when i halt the UML then only it is deleted.

Thanks and Best Regards, 
Adil Mujeeb



-----Original Message-----
From: user-mode-linux-user-admin@...
[mailto:user-mode-linux-user-admin@...]On Behalf Of
David Cannings
Sent: Friday, October 29, 2004 12:35 PM
To: user-mode-linux-user@...
Subject: Re: [uml-user] tap device


Adil Mujeeb, Noida wrote:
> Hi,
> I have some question related to tuntap & bridge.
> I have setup the bridge and started the UML. if i try to delete the tap
> device it says " resource or device is buzy"
> is there any way to delete the tap device while UML is running??

You will probably need to bring it down on the host, too.

# ifconfig mytap down

David


-------------------------------------------------------
This Newsletter Sponsored by: Macrovision 
For reliable Linux application installations, use the industry's leading
setup authoring tool, InstallShield X. Learn more and evaluate 
today. http://clk.atdmt.com/MSI/go/ins0030000001msi/direct/01/
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user@...
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user

RE: [uml-user] tap device

[Adil Mujeeb, Noida <adilm@hc...>]

i tried to down it on host and then try to delete it thru tunctl -t tap0 ,
it gives the same error message: TUNSETIFF Device or Resource is buzy.
Only when i halt the UML then only it is deleted.

Thanks and Best Regards, 
Adil Mujeeb



-----Original Message-----
From: user-mode-linux-user-admin@...
[mailto:user-mode-linux-user-admin@...]On Behalf Of
David Cannings
Sent: Friday, October 29, 2004 12:35 PM
To: user-mode-linux-user@...
Subject: Re: [uml-user] tap device


Adil Mujeeb, Noida wrote:
> Hi,
> I have some question related to tuntap & bridge.
> I have setup the bridge and started the UML. if i try to delete the tap
> device it says " resource or device is buzy"
> is there any way to delete the tap device while UML is running??

You will probably need to bring it down on the host, too.

# ifconfig mytap down

David


-------------------------------------------------------
This Newsletter Sponsored by: Macrovision 
For reliable Linux application installations, use the industry's leading
setup authoring tool, InstallShield X. Learn more and evaluate 
today. http://clk.atdmt.com/MSI/go/ins0030000001msi/direct/01/
_______________________________________________
User-mode-linux-user mailing list
User-mode-linux-user@...
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-user

RE: [uml-user] tap device

[Adil Mujeeb, Noida <adilm@hc...>]

hi,
i think when network interface goes down (ifconfig eth0 down) it closes the
corresponding tap device. So i guess i will be able to delete it.
 

Thanks and Best Regards,
Adil Mujeeb
-----Original Message-----
From: roland [mailto:for_spam@...]
Sent: Friday, October 29, 2004 1:08 PM
To: Adil Mujeeb, Noida; user-mode-linux-user@...;
user-mode-linux-devel-admin@...
Subject: Re: [uml-user] tap device



Hi Adil, 
>is there any way to delete the tap device while UML is running??
Why should somewone need to do this?
I think that the uml-process still keeps that ressource "occupied" -
regardless if eth0 inside uml is up or down.
 
regards
Roland
 

----- Original Message ----- 
From: Adil Mujeeb,  <mailto:adilm@...> Noida 
To: user-mode-linux-user@...
<mailto:user-mode-linux-user@...>  ;
user-mode-linux-devel-admin@...
<mailto:user-mode-linux-devel-admin@...>  
Sent: Friday, October 29, 2004 6:24 AM
Subject: [uml-user] tap device

Hi,
I have some question related to tuntap & bridge.
I have setup the bridge and started the UML. if i try to delete the tap
device it says " resource or device is buzy"
then i have down the UML interface device 'eth0' by `ifconfig eth0 down`
No if again try to to delete the tap device it says "resource or device is
buzy"
 
is there any way to delete the tap device while UML is running??

Thanks and Best Regards,
Adil Mujeeb

Re: [uml-user] VM's in UML

[William Hunt <wjh@hu...>]

original message from Sriram Ramarathnam follows:
>    I started a Virtual Machine in user mode linux using the following command
> ./control start
> and once it was completely booted i tried to start another VM using the
> same command, but the boot process got stuck at some point with the
> following message
> 
> INIT: Id "co" respawning too fast: diabled for 5 minutes
> INIT: no more processes left in this run level
> 
> and stays like that doesn't go ahead.
> 
> What does that mean ? what should i do to go ahead.?


These messages are generated by /sbin/init;

The first message says that some program as spawned
by the inittab line begining "co:2345:respawn:"
is repeatedly failing to execute or dying
much too soon after being spawned.

The program is probably /sbin/getty or similar,
probably trying to open a display device,
and probably failing due to bad device name,
nonexistant device, bad permissions on device, 
device already in use, or similar.

The second message says that, after disabling "co:",
there is nothing else to do and nowhere else to go.

So i would guess that the second VM is trying to get
a display device that is already in use by the first VM.



-- 
William Hunt, Portland Oregon USA

Re: [uml-user] tap device

[roland <for_spam@gm...>]

Hi Adil,=20
>is there any way to delete the tap device while UML is running??
Why should somewone need to do this?
I think that the uml-process still keeps that ressource "occupied" - =
regardless if eth0 inside uml is up or down.

regards
Roland

  ----- Original Message -----=20
  From: Adil Mujeeb, Noida=20
  To: user-mode-linux-user@... ; =
user-mode-linux-devel-admin@...=20
  Sent: Friday, October 29, 2004 6:24 AM
  Subject: [uml-user] tap device


  Hi,
  I have some question related to tuntap & bridge.
  I have setup the bridge and started the UML. if i try to delete the =
tap device it says " resource or device is buzy"
  then i have down the UML interface device 'eth0' by `ifconfig eth0 =
down`
  No if again try to to delete the tap device it says "resource or =
device is buzy"

  is there any way to delete the tap device while UML is running??
  Thanks and Best Regards,
  Adil Mujeeb

Re: [uml-user] tap device

[David Cannings <lists@ed...>]

Adil Mujeeb, Noida wrote:
> Hi,
> I have some question related to tuntap & bridge.
> I have setup the bridge and started the UML. if i try to delete the tap
> device it says " resource or device is buzy"
> is there any way to delete the tap device while UML is running??

You will probably need to bring it down on the host, too.

# ifconfig mytap down

David