Re: [Hampshire] Grub / boot issue

2010-09-28 Thread Vic

> It is pretty much as created and maintained by the Ubuntu install, so
> having
> removed the majority of the comments to keep it short:
>
> default 0
> timeout 3
> hiddenmenu
> ## ## End Default Options ##
>
> title   Ubuntu 8.04.4 LTS, kernel 2.6.24-21-server
> root(hd0,0)
> kernel  /boot/vmlinuz-2.6.24-21-server root=/dev/md0 ro quiet
> splash
> initrd  /boot/initrd.img-2.6.24-21-server
> quiet

OK, so it's unlikely to be a UUID problem.

The first thing I'd do is get rid of the "quiet" and "splash" options.
That might show you a bit more of what's going on. It might be an idea to
increase the timeout value as well - make sure there's enough time for the
screen to catch up.

I guess you've already checked that /dev/md0 is actually functional - if
the mirror is degraded, one of the partitions is unlikely to boot.

It's worth checking fdisk-s output there are no other partitions as well -
it's quite embarrassing to find you're not booting from the partition you
thought you were (not that I'd ever do that, oh no).

Errr - that's about all that occurs to me at present. Let us know if
anything comes from the above :-)

Vic.


--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread Keith Edmunds
On Tue, 28 Sep 2010 20:55:24 +0100, j...@jesoftware.freeserve.co.uk said:

> The easiest way to deal with the problem is to create a small partition
> on the drive which will fit s small linux distro.  Install into this
> partition.  When the small distro is running update grub.

Booting from a standalone CD (I favour SystemRescueCD) and using blkid(8)
and chroot(8) sounds considerably easier, not to say quicker.

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread John Cooper

On 28/09/10 20:55, john wrote:

The easiest way to deal with the problem is to create a small partition on the
drive which will fit s small linux distro.  Install into this partition.  When
the small distro is running update grub.

Latest Puppy should do it.

This will give you the details about the drives that you will need.

You then should be able to access the other linux.

Once the other linux is running you can remove the puppy linux.

This has saved me in the past.



Not quite sure what this will show/do as he said the 2nd disk will not 
boot. fdisk -l and blkid as root will give the drive details.


A simple procedure will show the disk labels/UUIDs

fdisk -l | grep "^/dev" | awk '{print $1}' | grep -v swap | grep -v Ext 
> /tmp/partlabels

for i in `cat /tmp/partlabels`
do
cmdlog "e2label $i"
cmdlog "dumpe2fs $i"
done
rm /tmp/partlabels




--
--
Discover Linux - Open Source Solutions to Business and Schools
http://discoverlinux.co.uk
--

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread john
The easiest way to deal with the problem is to create a small partition on the 
drive which will fit s small linux distro.  Install into this partition.  When 
the small distro is running update grub.

Latest Puppy should do it.

This will give you the details about the drives that you will need.

You then should be able to access the other linux.

Once the other linux is running you can remove the puppy linux.

This has saved me in the past.


On Tuesday 28 September 2010 10:15:38 Paul Tansom wrote:
> I must be missing something obvious on this, but I've done it before
> without problems and, although I've got a bit more investigating up my
> sleeve, this one is getting annoying!
> 
> I've got a system (Ubuntu 8.04 LTS Server) that is running quite happily on
> a software RAID mirror. One drive needs swapping, which I have just done
> without problems until I want to swap and replace the second one as well.
> Before I removed the failing drive I checked that I had grub installed in
> the boot sector of both drives and that I could boot off the good drive
> without the failing one connected - no problems whether connected to SATA1
> or SATA2 on the motherboard.
> 
> I installed the second drive, matched up the partition sizes, rebooted and
> resync'd the partitions. All is working fine - except...
> 
> If I remove the original drive to boot purely from the new one (as I did
> initially for the first swap out) I cannot boot. Grub has been installed in
> the boot sector, but somewhere there must be a direct reference to the
> original drive with some form of identification. I just can't think what!
> 
> If the drive is connected to SATA1 the boot hangs after the grub menu with:
> 
> Loading please wait...
> 
> at the top of the screen, and at the bottom:
> 
> Kernel alive
> kernel direct mapping tables up to 1 @ 8000-d000
> 
> Which is exactly what you see during a successful boot.
> 
> If the drive is connected to SATA2 the boot fails with
> 
> GRUB Hard Disk Error
> 
> Clearly it is looking for something that has to be on SATA1 and seems to
> want the original drive. Oddly the original drive still boots on either
> SATA1 or SATA2 without problems. The entire disk is sync'd with the RAID
> mirror so the drive contents are the same, and grub has been installed in
> the same way to both drives. Both drives are of the same capacity (500G
> and amazingly the capacity matches exactly, not just in marketing termes),
> although the original is Samsung and the new one is WD.
> 
> Anyone have any ideas? I'm about to double check the location of
> /boot/grub/stage1 at the grub boot stage in case that is different. There's
> half a temptation to install 3 drives at the same time to sync up the new
> drive, but I'm not happy to do that until I can boot cleanly of the new
> drive on its own.
> 
> This is annoying as I've done the same process before (admittedly on PATA
> not SATA) and even done it to higher capaicty drives and then grown the
> partitions to use the extra space.

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


[Hampshire] [ADMIN] October meeting

2010-09-28 Thread Hants LUG Chairman
Hi,

This is your mid-week reminder that the next LUG meeting and AGM will take 
place on Saturday 2 October at Southampton University between the normal hours 
of 10:00 and 4:30. We have one talk already and it will be our AGM.

So far there have been no nominations on the list for the committee but I am 
aware of one off-list nomination for the chairman's position. Please send your 
nominations to this list soon!

### In other news

The Surrey LUG are having their October meeting at Nokia as Bob recently 
reported so if you miss our meeting there is a Surrey one a week later.

Though not confirmed it's hoped that the November meeting may be a joint 
Surrey/Hants LUG meeting at Red Hat.

On 27 November there is the Barcamp meeting in Southampton. 
http://barcampsoton.eventbrite.com/

On 04 December there is the annual London Perl Workshop in London. 
http://conferences.yapceurope.org/lpw2010/

-- 
Adam Trickett
  Chairman, Hampshire Linux Users Group
http://www.hants.lug.org.uk/


signature.asc
Description: This is a digitally signed message part.
--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--

Re: [Hampshire] [ADMIN] AGM and next meeting

2010-09-28 Thread Victor Churchill
On 25 September 2010 09:01, Hants LUG Chairman
 wrote:

>
> If you would like to give a talk at the meeting then please let us know

I will offer the talk that did not happen at Hursley, on beginning
cloud computing with AWS (provided, that is, that my lapdog will talk
to the OHP ;-).

Also: is it still the case that people who have been there before do
not need to pre-register their wireless MAC addresses?


--
regards,

Victor Churchill,
Bournemouth

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


[Hampshire] OpenOffice Windows vs Linux

2010-09-28 Thread Bob Dunlop
Hi,

Someone at work has a problem and I wondered if anyone could give me some
hints to resolve it for him.

He has a small OpenOffice macro package (written by his wife so we can't
change that) which provides keystokes combinations for things like umlauts.

The package was generated and works with OpenOffice under Windows, but we
can't figure out how to install/get it working with OpenOffice under Linux.

Can anyone give me a hint as to where I should start looking ?  I'm not
a regular OpenOffice user so assume nothing.

-- 
Bob Dunlop

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


[Hampshire] FW: Reminder : BaB @ Nokia on 9 Oct 2010 : 11am - 5pm'ish

2010-09-28 Thread robert.beattie

 October !



--
Bob Beattie
Senior Technical Support Engineer
Camera Development Systems, MP R&D,
Nokia Southwood, UK
Tel : +44 (0)1252 866452
www.nokia.com
--

-Original Message-
From: Beattie Robert (Nokia-MP/Southwood) 
Sent: 28 September 2010 15:57
To: 'General Linux/Unix community List'
Cc: 'Hampshire LUG Discussion List'
Subject: Reminder : BaB @ Nokia on 9 Sep 2010 : 11am - 5pm'ish

Surrey LUG Bring-a-Box reminder :  Nokia, GU14 0NG

As usual, please email me next-of-kin details so that our security people are 
kept happy. :) Mark the subject as "Linux meeting".
Fellow Hants Luggers also welcome.

--
Bob Beattie
Senior Technical Support Engineer
Camera Development Systems, MP R&D,
Nokia Southwood, UK
Tel : +44 (0)1252 866452
www.nokia.com
--


--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


[Hampshire] Reminder : BaB @ Nokia on 9 Sep 2010 : 11am - 5pm'ish

2010-09-28 Thread robert.beattie
Surrey LUG Bring-a-Box reminder :  Nokia, GU14 0NG

As usual, please email me next-of-kin details so that our security people are 
kept happy. :)
Mark the subject as "Linux meeting".
Fellow Hants Luggers also welcome.

--
Bob Beattie
Senior Technical Support Engineer
Camera Development Systems, MP R&D,
Nokia Southwood, UK
Tel : +44 (0)1252 866452
www.nokia.com
--


--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread Paul Tansom
** Vic  [2010-09-28 12:08]:
> > Paul, is this a UUID problem.
> 
> I was just about to suggest the same thing.
> 
> Please post your grub.conf (or menu.lst) If there are any UUID entries in
> there, you'll ned to change them. I tend to use volume labels instead for
> this sort of thing - it makes swapping out drives a little easier...
** end quote [Vic]

It is pretty much as created and maintained by the Ubuntu install, so having
removed the majority of the comments to keep it short:

default 0
timeout 3
hiddenmenu
## ## End Default Options ##

title   Ubuntu 8.04.4 LTS, kernel 2.6.24-21-server
root(hd0,0)
kernel  /boot/vmlinuz-2.6.24-21-server root=/dev/md0 ro quiet splash
initrd  /boot/initrd.img-2.6.24-21-server
quiet

title   Ubuntu 8.04.4 LTS, kernel 2.6.24-21-server (recovery mode)
root(hd0,0)
kernel  /boot/vmlinuz-2.6.24-21-server root=/dev/md0 ro single
initrd  /boot/initrd.img-2.6.24-21-server

title   Ubuntu 8.04.4 LTS, kernel 2.6.24-16-server
root(hd0,0)
kernel  /boot/vmlinuz-2.6.24-16-server root=/dev/md0 ro quiet splash
initrd  /boot/initrd.img-2.6.24-16-server
quiet
title   Ubuntu 8.04.4 LTS, kernel 2.6.24-16-server (recovery mode)
root(hd0,0)
kernel  /boot/vmlinuz-2.6.24-16-server root=/dev/md0 ro single
initrd  /boot/initrd.img-2.6.24-16-server

title   Ubuntu 8.04.4 LTS, memtest86+
root(hd0,0)
kernel  /boot/memtest86+.bin
quiet

### END DEBIAN AUTOMAGIC KERNELS LIST

title   Ubuntu 8.04.4 LTS, kernel 2.6.24-21-server
root(hd1,0)
kernel  /boot/vmlinuz-2.6.24-21-server root=/dev/md0 ro quiet splash
initrd  /boot/initrd.img-2.6.24-21-server
quiet

title   Ubuntu 8.04.4 LTS, kernel 2.6.24-21-server (recovery mode)
root(hd1,0)
kernel  /boot/vmlinuz-2.6.24-21-server root=/dev/md0 ro single
initrd  /boot/initrd.img-2.6.24-21-server

title   Ubuntu 8.04.4 LTS, memtest86+
root(hd1,0)
kernel  /boot/memtest86+.bin
quiet


Note the last three entries allow me to use the BIOS boot menu and boot from
the second HD if the first is failing or blank. The sda (hd0,0) is the replaced
drive.

-- 
Paul Tansom | Aptanet Ltd. | http://www.aptanet.com/ | 023 9238 0001
==
Registered in England  |  Company No: 4905028  |  Registered Office:
Crawford House, Hambledon Road, Denmead, Waterlooville, Hants, PO7 6NU

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread Vic

> Paul, is this a UUID problem.

I was just about to suggest the same thing.

Please post your grub.conf (or menu.lst) If there are any UUID entries in
there, you'll ned to change them. I tend to use volume labels instead for
this sort of thing - it makes swapping out drives a little easier...

Vic.


--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread Paul Tansom
** John Cooper  [2010-09-28 10:48]:
> On 28/09/10 10:15, Paul Tansom wrote:
>> I must be missing something obvious on this, but I've done it before without
>> problems and, although I've got a bit more investigating up my sleeve, this 
>> one
>> is getting annoying!
>>
>> I've got a system (Ubuntu 8.04 LTS Server) that is running quite happily on a
>> software RAID mirror. One drive needs swapping, which I have just done 
>> without
>> problems until I want to swap and replace the second one as well. Before I
>> removed the failing drive I checked that I had grub installed in the boot
>> sector of both drives and that I could boot off the good drive without the
>> failing one connected - no problems whether connected to SATA1 or SATA2 on 
>> the
>> motherboard.
>>
>> I installed the second drive, matched up the partition sizes, rebooted and
>> resync'd the partitions. All is working fine - except...
>>
>> If I remove the original drive to boot purely from the new one (as I did
>> initially for the first swap out) I cannot boot. Grub has been installed in 
>> the
>> boot sector, but somewhere there must be a direct reference to the original
>> drive with some form of identification. I just can't think what!
>>
>> If the drive is connected to SATA1 the boot hangs after the grub menu with:
>>
>> Loading please wait...
>>
>> at the top of the screen, and at the bottom:
>>
>> Kernel alive
>> kernel direct mapping tables up to 1 @ 8000-d000
>>
>> Which is exactly what you see during a successful boot.
>>
>> If the drive is connected to SATA2 the boot fails with
>>
>> GRUB Hard Disk Error
>>
>> Clearly it is looking for something that has to be on SATA1 and seems to want
>> the original drive. Oddly the original drive still boots on either SATA1 or
>> SATA2 without problems. The entire disk is sync'd with the RAID mirror so the
>> drive contents are the same, and grub has been installed in the same way to
>> both drives. Both drives are of the same capacity (500G and amazingly the
>> capacity matches exactly, not just in marketing termes), although the 
>> original
>> is Samsung and the new one is WD.
>>
>> Anyone have any ideas? I'm about to double check the location of
>> /boot/grub/stage1 at the grub boot stage in case that is different. There's
>> half a temptation to install 3 drives at the same time to sync up the new
>> drive, but I'm not happy to do that until I can boot cleanly of the new drive
>> on its own.
>>
>> This is annoying as I've done the same process before (admittedly on PATA not
>> SATA) and even done it to higher capaicty drives and then grown the 
>> partitions
>> to use the extra space.
>>
>
> Paul, is this a UUID problem. Both grub and fstab can state unique UUIDs  
> and LVM names (fstab also can use disk label names). My grub has
>
> rd_LUKS_UUID=luks-411e43ee----  
> rd_LVM_LV=vg_mail/lv_root rd_LVM_LV=vg_mail/lv_swap
>
> in the kernel line. fstab has root partition as  
> /dev/mapper/vg_mail-lv_root amd /boot as UUID=X
** end quote [John Cooper]

That was my thought, but as yet I've found no way to check the UUID in use or 
change it.

/boot is on the root partition, not on separate one. The UUID in fstab refers
to /dev/md0, so I would assume that even if one of the drives used for this
mirror changes the UUID should remain the same. the grub menu.lst makes no
reference to UUID, and having run grub-update and install-grub /dev/sda and the
same for sdb I would hope that any changes to UUID have been accounted for.

Those top two lines from grub don't look familiar at all in format. All my
references are /dev/mdx ones. I was looking to invextigate any issues with the
groot values, or possibly the updatedefaultentry perhaps. I need to read up a
bit more.


-- 
Paul Tansom | Aptanet Ltd. | http://www.aptanet.com/ | 023 9238 0001
==
Registered in England  |  Company No: 4905028  |  Registered Office:
Crawford House, Hambledon Road, Denmead, Waterlooville, Hants, PO7 6NU

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


Re: [Hampshire] Grub / boot issue

2010-09-28 Thread John Cooper

On 28/09/10 10:15, Paul Tansom wrote:

I must be missing something obvious on this, but I've done it before without
problems and, although I've got a bit more investigating up my sleeve, this one
is getting annoying!

I've got a system (Ubuntu 8.04 LTS Server) that is running quite happily on a
software RAID mirror. One drive needs swapping, which I have just done without
problems until I want to swap and replace the second one as well. Before I
removed the failing drive I checked that I had grub installed in the boot
sector of both drives and that I could boot off the good drive without the
failing one connected - no problems whether connected to SATA1 or SATA2 on the
motherboard.

I installed the second drive, matched up the partition sizes, rebooted and
resync'd the partitions. All is working fine - except...

If I remove the original drive to boot purely from the new one (as I did
initially for the first swap out) I cannot boot. Grub has been installed in the
boot sector, but somewhere there must be a direct reference to the original
drive with some form of identification. I just can't think what!

If the drive is connected to SATA1 the boot hangs after the grub menu with:

Loading please wait...

at the top of the screen, and at the bottom:

Kernel alive
kernel direct mapping tables up to 1 @ 8000-d000

Which is exactly what you see during a successful boot.

If the drive is connected to SATA2 the boot fails with

GRUB Hard Disk Error

Clearly it is looking for something that has to be on SATA1 and seems to want
the original drive. Oddly the original drive still boots on either SATA1 or
SATA2 without problems. The entire disk is sync'd with the RAID mirror so the
drive contents are the same, and grub has been installed in the same way to
both drives. Both drives are of the same capacity (500G and amazingly the
capacity matches exactly, not just in marketing termes), although the original
is Samsung and the new one is WD.

Anyone have any ideas? I'm about to double check the location of
/boot/grub/stage1 at the grub boot stage in case that is different. There's
half a temptation to install 3 drives at the same time to sync up the new
drive, but I'm not happy to do that until I can boot cleanly of the new drive
on its own.

This is annoying as I've done the same process before (admittedly on PATA not
SATA) and even done it to higher capaicty drives and then grown the partitions
to use the extra space.



Paul, is this a UUID problem. Both grub and fstab can state unique UUIDs 
and LVM names (fstab also can use disk label names). My grub has


rd_LUKS_UUID=luks-411e43ee---- 
rd_LVM_LV=vg_mail/lv_root rd_LVM_LV=vg_mail/lv_swap


in the kernel line. fstab has root partition as 
/dev/mapper/vg_mail-lv_root amd /boot as UUID=X



John.
--
--
Discover Linux - Open Source Solutions to Business and Schools
http://discoverlinux.co.uk
--

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--


[Hampshire] Grub / boot issue

2010-09-28 Thread Paul Tansom
I must be missing something obvious on this, but I've done it before without
problems and, although I've got a bit more investigating up my sleeve, this one
is getting annoying!

I've got a system (Ubuntu 8.04 LTS Server) that is running quite happily on a
software RAID mirror. One drive needs swapping, which I have just done without
problems until I want to swap and replace the second one as well. Before I
removed the failing drive I checked that I had grub installed in the boot
sector of both drives and that I could boot off the good drive without the
failing one connected - no problems whether connected to SATA1 or SATA2 on the
motherboard.

I installed the second drive, matched up the partition sizes, rebooted and
resync'd the partitions. All is working fine - except...

If I remove the original drive to boot purely from the new one (as I did
initially for the first swap out) I cannot boot. Grub has been installed in the
boot sector, but somewhere there must be a direct reference to the original
drive with some form of identification. I just can't think what!

If the drive is connected to SATA1 the boot hangs after the grub menu with:

Loading please wait...

at the top of the screen, and at the bottom:

Kernel alive
kernel direct mapping tables up to 1 @ 8000-d000

Which is exactly what you see during a successful boot.

If the drive is connected to SATA2 the boot fails with

GRUB Hard Disk Error

Clearly it is looking for something that has to be on SATA1 and seems to want
the original drive. Oddly the original drive still boots on either SATA1 or
SATA2 without problems. The entire disk is sync'd with the RAID mirror so the
drive contents are the same, and grub has been installed in the same way to
both drives. Both drives are of the same capacity (500G and amazingly the
capacity matches exactly, not just in marketing termes), although the original
is Samsung and the new one is WD.

Anyone have any ideas? I'm about to double check the location of
/boot/grub/stage1 at the grub boot stage in case that is different. There's
half a temptation to install 3 drives at the same time to sync up the new
drive, but I'm not happy to do that until I can boot cleanly of the new drive
on its own.

This is annoying as I've done the same process before (admittedly on PATA not
SATA) and even done it to higher capaicty drives and then grown the partitions
to use the extra space.

-- 
Paul Tansom | Aptanet Ltd. | http://www.aptanet.com/ | 023 9238 0001
==
Registered in England  |  Company No: 4905028  |  Registered Office:
Crawford House, Hambledon Road, Denmead, Waterlooville, Hants, PO7 6NU

--
Please post to: Hampshire@mailman.lug.org.uk
Web Interface: https://mailman.lug.org.uk/mailman/listinfo/hampshire
LUG URL: http://www.hantslug.org.uk
--