Re: FATAL ERROR. Filesystem UNSUPPORTED not supported for initrd image. Terminating.

Adi Spivak said on Thu, Feb 10, 2011 at 11:39:56AM +0200:

> Mindi Linux mini-distro generator v2.0.4-r2045

Please update at ftp://ftp.mondorescue.org
It's a known and now old issue.

> Mindi-BusyBox v1.2.1 (2009.06.02-07:09+0000) multi-call binary

Update as well.

Bruno.
--

-- 
Open Source & Linux Profession Lead EMEA           / http://opensource.hp.com
HP/Intel/Red Hat Open Source Solutions Initiative  / http://www.hpintelco.net
http://www.HyPer-Linux.org  http://mondorescue.org http://project-builder.org
La musique ancienne?  http://www.musique-ancienne.org http://www.medieval.org

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: FATAL ERROR. Filesystem UNSUPPORTED not supported for initrd image. Terminating.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

_______________________________________________
Mondo-devel mailing list
Mondo-devel <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mondo-devel

Re: mondorestore to KVM - please help

Hi

There is an error in the URLs 
http://localhost...

mindi uses busybox to provide a light Linux environment at restore time, so it's possible to shift to a
busybox shell through Alt-F2 (Alt-F1 will allow you later to return to mondorestore GUI screen), then to
save mondorestore.log you should be able to tftp it to a tftp server, because tftp is embedded in mindi-busybox.

Rgds.

-----Original Message-----
From: david [mailto:david <at> kenpro.com.au] 
Sent: vendredi 11 février 2011 01:09
To: Mondo mailing list
Subject: [Mondo-devel] mondorestore to KVM - please help

Has anyone succeeded in doing this?

The last 3 pages of the log is here: 
http://localhost/david/mondo/110211/mondo_log_last3pages.jpg
The screenshot where it fails is here:
http://localhost/david/mondo/110211/mondoTerminatingResult254.png

I can't find a good way to save the mondo log while doing this restore 
because I never get a mountable file system to save it to.

I've noticed that "running klogd" takes a very long time - 5 minutes - 
is that normal?

The same mondoarchive restores to VirtualBox, so the fault is not with 
the archive itself.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
_______________________________________________
Mondo-devel mailing list
Mondo-devel <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mondo-devel

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: mondorestore to KVM - please help

Did you checked http://www.mondorescue.org/docs/p2v.pdf ?

My 2 cents. 

-----Original Message-----
From: Gattegno, Victor (GCC - Software) 
Sent: vendredi 11 février 2011 10:29
To: 'Mondo mailing list'
Subject: RE: [Mondo-devel] mondorestore to KVM - please help

Hi

There is an error in the URLs 
http://localhost...

mindi uses busybox to provide a light Linux environment at restore time, so it's possible to shift to a
busybox shell through Alt-F2 (Alt-F1 will allow you later to return to mondorestore GUI screen), then to
save mondorestore.log you should be able to tftp it to a tftp server, because tftp is embedded in mindi-busybox.

Rgds.

-----Original Message-----
From: david [mailto:david <at> kenpro.com.au] 
Sent: vendredi 11 février 2011 01:09
To: Mondo mailing list
Subject: [Mondo-devel] mondorestore to KVM - please help

Has anyone succeeded in doing this?

The last 3 pages of the log is here: 
http://localhost/david/mondo/110211/mondo_log_last3pages.jpg
The screenshot where it fails is here:
http://localhost/david/mondo/110211/mondoTerminatingResult254.png

I can't find a good way to save the mondo log while doing this restore 
because I never get a mountable file system to save it to.

I've noticed that "running klogd" takes a very long time - 5 minutes - 
is that normal?

The same mondoarchive restores to VirtualBox, so the fault is not with 
the archive itself.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb
_______________________________________________
Mondo-devel mailing list
Mondo-devel <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mondo-devel

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: Debian 6.0 - Squeeze

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

_______________________________________________
Mondo-devel mailing list
Mondo-devel <at> lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/mondo-devel

Re: Debian 6.0 - Squeeze

Hi Edward,
Thanks for your help, but for what I understand the bios_grub flag ang
biosgub partition type are only available with GPT partition tables,
but I'm not in that case.

What I've done is modifying the number of sectors/track with fdisk to
63 (see fdisk output below), with that my first partition start on the
sector 63 and grub2 is fine with that
But If I redo the whole cloning process (archive + restore), the
problem reappear again with the cloned system, grub-MR ran just fine
(see output below) but I cannot boot the system. So I boot my restore
CD again, go to the shell, mount the restoring partition, ran
grub-install + update-grub and then I can boot.
So, I think there is always a problem with the grub-MR script when we
use grub2. I have also an issue with the fstab file, populated with
UUID.

Other minor problems :
- mondoarchive doesn't recognize my keymap, I don't understand why, It
worked well with older version.
- mondorestore is in ascii text mode (no blue background, etc ...)

Rémi

/mondo/grub-MR
--------------------------------end of output------------------------------
...ran just fine. 
[Main] mondo-rstr-tools.c-≥run_grub#1527: Yay! grub-MR found...
[Main] mondo-rstr-tools.c-≥run_grub#1529: command = grub-MR
/dev/cciss/c0d0 /tmp/mountlist.txt
Running GRUB...
        [Main] mondo-rstr-tools.c-≥run_grub#1584: grub-MR
/dev/cciss/c0d0 /tmp/mountlist.txt
Now I'll use grub-install
Installation finished. No error reported.
grub-install returned 0
running: grub-MR /dev/cciss/c0d0 /tmp/mountlist.txt >
//mondo.tmp.W80kct/mondo-run-prog-thing.tmp 2>
//mondo.tmp.W80kct/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
Installation finished. No error reported.
--------------------------------end of output------------------------------
...ran just fine. 
Done.

# fdisk -lu
Disk /dev/cciss/c0d0: 18.2 GB, 18203197440 bytes
255 heads, 63 sectors/track, 2213 cylinders, total 35553120 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x0009a68f

           Device Boot      Start         End      Blocks   Id  System
/dev/cciss/c0d0p1   *          63    11872034     5935986   83  Linux
/dev/cciss/c0d0p2        11872035    30844799     9486382+  83  Linux
/dev/cciss/c0d0p3        30844800    35551844     2353522+  82  Linux
swap / Solaris

2011/2/11 Edward J. Ouellette <ejouellette <at> gmail.com>:
> Hi Rémi,
> Glad my suggestions helped :)
> Instead of creating a "fake blank partition" at the start of the drive,
> you can create a small partition (on my machine it's only 128KB),
> and set partition type = biosgrub. I labeled mine "grub_boot".
>
> Usually this partition _should_ be the first one (31KB / 64KB / 128KB in
> size)
> and it should be flaged with the bios_grub flag. (parted <device> set 1
> bios_grub on)
>
> Then grub2 will be happy, and mondo should restore the biosgrub partition
> correctly...
> Hope this helps!
> Ed
>
>
> On Thu, Feb 10, 2011 at 12:36 PM, Rémi <mirebob <at> gmail.com> wrote:
>>
>> Hi Edward,
>>
>> > The newer releases of GRUB2 need a special partition to store the
>> > core.img
>> > (this is the kernel image of grub + necessary modules for reading your
>> > filesystem). Former releases used the so called 62 block gap between the
>> > mbr
>> > (block 1) and the first partition block 63.
>>
>> yes I just read some stuff about that (including the link you
>> mentioned), and I understood it was not a mondo issue, nor a grub2 but
>> a disk/partition issue and the problem is also present on the original
>> system
>> the gap before my first partition was only 32 blocks (see fdisk output
>> below), not enough for grub2, so I resize this partition to increase
>> the gap, everything works fine now
>>
>> but If I redo the whole cloning process (archive + restore), the
>> problem reappear because mondo part my disk again and don't care about
>> the gap before the first partition, so I think I must create a fake
>> blank first partition but I really don't like that idea
>>
>> btw, I know where the problem is now, thanks all for your answers
>>
>>
>> # fdisk -lu
>> Disk /dev/cciss/c0d0: 18.2 GB, 18203197440 bytes
>> 255 heads, 32 sectors/track, 4357 cylinders, total 35553120 sectors
>> Units = sectors of 1 * 512 = 512 bytes
>> Sector size (logical/physical): 512 bytes / 512 bytes
>> I/O size (minimum/optimal): 512 bytes / 512 bytes
>> Disk identifier: 0x00031c20
>>
>>           Device Boot      Start         End      Blocks   Id  System
>> /dev/cciss/c0d0p1   *          32    10175519     5087744   83  Linux
>> /dev/cciss/c0d0p2        10175520    15299999     2562240   82  Linux
>> swap / Solaris
>> /dev/cciss/c0d0p3        15300000    35553119    10126560   83  Linux
>>
>>
>> Rémi
>>
>>
>> ------------------------------------------------------------------------------
>> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
>> Pinpoint memory and threading errors before they happen.
>> Find and fix more than 250 security defects in the development cycle.
>> Locate bottlenecks in serial and parallel code that limit performance.
>> http://p.sf.net/sfu/intel-dev2devfeb
>> _______________________________________________
>> Mondo-devel mailing list
>> Mondo-devel <at> lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/mondo-devel
>
>
> ------------------------------------------------------------------------------
> The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
> Pinpoint memory and threading errors before they happen.
> Find and fix more than 250 security defects in the development cycle.
> Locate bottlenecks in serial and parallel code that limit performance.
> http://p.sf.net/sfu/intel-dev2devfeb
> _______________________________________________
> Mondo-devel mailing list
> Mondo-devel <at> lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/mondo-devel
>
>

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: Ubuntu 10.10 and Mondo are no friends anymore

Gattegno, Victor (GCC - Software) wrote:
> 
> 
>>From your "uname -a", your kernel version is 2.6.35-26
> Is it a vanilla (test) kernel ?
> I just upgraded my Ubuntu 10.10 and my kernel version is 2.6.35-25 ...
> 
I've got pre-release and unsupported updates enabled in Ubuntu
update-manager. Maybe that's why we're on different kernel versions

> To check software versions, execute :
> dpkg -l | egrep 'afio|buffer|glibc|newt' | sort
> 
ii  afio                                  2.5-5                                                     
archive file manipulation program
ii  buffer                                1.19-10                                                   
Buffering/reblocking program for tape backups, printing, etc.
ii  libdirectfb-1.2-9                     1.2.10.0-4ubuntu2                                         
direct frame buffer graphics - shared libraries
ii  libnewt0.52                           0.52.11-1                                                 
Not Erik's Windowing Toolkit - text mode windowing with slang
ii  libprotobuf6                          2.3.0-2ubuntu1                                            
protocol buffers C++ library
ii  libprotoc6                            2.3.0-2ubuntu1                                            
protocol buffers compiler library
ii  libxine1-console                      1.1.18.1-4ubuntu4                                         
libaa/libcaca/framebuffer/directfb related plugins for libxine1
ii  protobuf-compiler                     2.3.0-2ubuntu1                                            
compiler for protocol buffer definition files
ii  python-newt                           0.52.11-1                                                 
A NEWT module for Python
ii  python-protobuf                       2.3.0-2ubuntu1                                            
Python bindings for protocol buffers
rc  libdirectfb-1.2-0                     1.2.8-5ubuntu2                                            
direct frame buffer graphics - shared libraries
rc  libprotobuf5                          2.2.0a-0.1ubuntu1                                         
protocol buffers C++ library
rc  libprotoc5                            2.2.0a-0.1ubuntu1                                         
protocol buffers compiler library

> Without the /var/log/mondoarchive.log content, it's difficult to see
> what's happening, please (compress it and) post it to the mailing list.
> 
[TH=16058] libmondo-tools.c-≥reset_bkpinfo#753: Hi
root is mounted at /dev/sda

No, Schlomo, that doesn't mean /dev/sda is the root partition. It's just a
debugging message. Relax. It's part of am_I_in_disaster_recovery_mode().
[TH=16058] libmondo-devices.c-≥am_I_in_disaster_recovery_mode#158: Is this a
ramdisk? result = 0
        [TH=16058] libmondo-tools.c-≥setup_tmpdir#737: Failed to create
global tmp directory /mnt/hgfs/mondo.tmp.fg2OVd for Mondo.
running: umount /mnt/cdrom > /mondo-run-prog-thing.tmp 2>
/mondo-run-prog-thing.err
--------------------------------start of output-----------------------------
umount: /mnt/cdrom: not mounted
--------------------------------end of output------------------------------
...ran with res=256

That's all. I deleted mondoarchive.log before this, so I'm sure this is the
only bit being written to mondoarchive.log

Hmmm...  "Failed to create global tmp directory /mnt/hgfs/mondo.tmp.fg2OVd
for Mondo.". Why is it trying to create a directory there? That's obviously
not going to work (/mnt/hgfs is a virtual filesystem from VMWare that let's
you access shared directories on your host system).

> Did you checked mondorescue FAQ ?
> http://trac.mondorescue.org/wiki/FAQ
> Q18/ Why does mindi seg fault on Ubuntu?
> 
Tried that, doesn't make a difference.

> About upgrades and wrong mondo version, check also
> Q11/ Does mondo work with Debian/Ubuntu distributions?
> 
I don't think this applies here. I'm sure I have installed the package
versions from the mondorescue FTP site.

Regards, Maarten
--

-- 
View this message in context: http://mondorescue-mailing-list.679749.n3.nabble.com/Ubuntu-10-10-and-Mondo-are-no-friends-anymore-tp2438692p2479112.html
Sent from the MondoRescue Mailing List mailing list archive at Nabble.com.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: Ubuntu 10.10 and Mondo are no friends anymore

In case it might make a difference, here is my backup script. It's not clear
to me why mondoarchive would want to make a temp directory under /mnt/hgfs
(see my previous mail).

#!/bin/bash

SCRATCH=/tmp/mondoarchive.scratch.$$
TEMPDIR=/tmp/mondoarchive.temp.$$

EXCLUDE=""
EXCLUDE="$EXCLUDE /home/boekhold/Downloads"
EXCLUDE="$EXCLUDE /mnt/hgfs"
EXCLUDE="$EXCLUDE $SCRATCH"
EXCLUDE="$EXCLUDE $TEMPDIR"

DATE=$(date '+%Y%m%d')
DIR=/mnt/hgfs/C/Maarten/Backups/mondorescue-$DATE

mkdir -p $DIR
mkdir -p $SCRATCH
mkdir -p $TEMPDIR

# -O            : backup
# -i            : create ISO image
# -s 4480m      : decent size for a DVD image
# -3            : average compression
# -d $DIR       : where to store the images
# -E "$EXCLUDE" : list of directories to exclude from the backup
# -D $SCRATCH   : scratch dir where it prepares the images
# -T $TEMPDIR   : temp dir for non-iso temporary files

mondoarchive\
        -O \
        -i \
        -s 4480m \
        -3 \
        -d $DIR \
        -E "$EXCLUDE" \
        -S $SCRATCH \
        -T $TEMPDIR

rm -rf $SCRATCH $TEMPDIR
--

-- 
View this message in context: http://mondorescue-mailing-list.679749.n3.nabble.com/Ubuntu-10-10-and-Mondo-are-no-friends-anymore-tp2438692p2479122.html
Sent from the MondoRescue Mailing List mailing list archive at Nabble.com.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: Ubuntu 10.10 and Mondo are no friends anymore

boekhold said on Sat, Feb 12, 2011 at 12:40:44AM -0800:

> > Without the /var/log/mondoarchive.log content, it's difficult to see
> > what's happening, please (compress it and) post it to the mailing list.
> > 
> [TH=16058] libmondo-tools.c-≥reset_bkpinfo#753: Hi
> root is mounted at /dev/sda
> 
> No, Schlomo, that doesn't mean /dev/sda is the root partition. It's just a
> debugging message. Relax. It's part of am_I_in_disaster_recovery_mode().
> [TH=16058] libmondo-devices.c-≥am_I_in_disaster_recovery_mode#158: Is this a
> ramdisk? result = 0
>         [TH=16058] libmondo-tools.c-≥setup_tmpdir#737: Failed to create
> global tmp directory /mnt/hgfs/mondo.tmp.fg2OVd for Mondo.

That's why we always need the logs 

> Hmmm...  "Failed to create global tmp directory /mnt/hgfs/mondo.tmp.fg2OVd
> for Mondo.". Why is it trying to create a directory there? That's obviously
> not going to work (/mnt/hgfs is a virtual filesystem from VMWare that let's
> you access shared directories on your host system).

Because there is a function in mondoarchive which originally (before
your -T option analysis) tries to setup a temp dir, and it found it was
the most appropriate place for you (larger FS).

Could you give us the result of df -T please ?
I need to exclude it from the possible FS to look for.

As a workaround, you can setup the env var TMPDIR or TMP to avoid that
issue. Could be worth documenting on the wiki as well then.

Bruno.
--

-- 
Open Source & Linux Profession Lead EMEA           / http://opensource.hp.com
HP/Intel/Red Hat Open Source Solutions Initiative  / http://www.hpintelco.net
http://www.HyPer-Linux.org  http://mondorescue.org http://project-builder.org
La musique ancienne?  http://www.musique-ancienne.org http://www.medieval.org

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb

Re: Ubuntu 10.10 and Mondo are no friends anymore

Bruno Cornec-2 wrote:
> 
> 
> Because there is a function in mondoarchive which originally (before
> your -T option analysis) tries to setup a temp dir, and it found it was
> the most appropriate place for you (larger FS).
> 
> Could you give us the result of df -T please ?
> I need to exclude it from the possible FS to look for.
> 

root <at> boekhold:~/local/script# df -T
Filesystem    Type   1K-blocks      Used Available Use% Mounted on
/dev/sda1     ext4    79102432  18112648  56971528  25% /
none      devtmpfs      765836       212    765624   1% /dev
none         tmpfs      771568      1260    770308   1% /dev/shm
none         tmpfs      771568       224    771344   1% /var/run
none         tmpfs      771568         0    771568   0% /var/lock
none         tmpfs      771568         0    771568   0% /lib/init/rw
.host:/     vmhgfs   244196348 105563416 138632932  44% /mnt/hgfs

I must say that I find it weird behaviour that a program would ignore an
explicitly specified path in favour of an automatically determined
location... But I'm sure you had good reasons for this.

> As a workaround, you can setup the env var TMPDIR or TMP to avoid that
> issue. Could be worth documenting on the wiki as well then.
> 

OK, done that. After recreating /tmp (my backup script contained a "rm -rf
$TEMPDIR" at the end, so when I changed TEMPDIR=/tmp/mondo with TEMPDIR=/tmp
it kinda screwed things up :)) and changing my fstab from using UUID tp
/dev/sdaX entries it's running.

About those UUIDs btw. Is there a bug in mindi? The mindi log file says it
can't resolve those UUIDs and that I need to install findfs/blkid/vol_id.
The first 2 are available on my system, yet mindi doesn't seem to use them.

--

-- 
View this message in context: http://mondorescue-mailing-list.679749.n3.nabble.com/Ubuntu-10-10-and-Mondo-are-no-friends-anymore-tp2438692p2479240.html
Sent from the MondoRescue Mailing List mailing list archive at Nabble.com.

------------------------------------------------------------------------------
The ultimate all-in-one performance toolkit: Intel(R) Parallel Studio XE:
Pinpoint memory and threading errors before they happen.
Find and fix more than 250 security defects in the development cycle.
Locate bottlenecks in serial and parallel code that limit performance.
http://p.sf.net/sfu/intel-dev2devfeb