[linux-lvm] First try, first fail with LVM.

[linux-lvm] First try, first fail with LVM.

[Roger Eriksson]

I'm new at this list and new with linux lvm.

The story this far. I bought a new hard disk to my file server at home and
decided to reinstall the thing with a newer version of Linux (Red Hat 6.2
before, now Mandrake 9.2). I decided to go with LVM and reiserfs, during the
setup of Mandrake 9.2.

On the server I have two sets of disks. Three 4.6 Gb SCSI disks that I
joined together using LVM and three larger IDE disks that I also joined
together. Everything went well and I started to copy back the files that I
earlier copied to my Window and Linux desktop computers. No problems with
that and since it got late a set the computers to move the files back and
went to bed. The next morning when I checked the damn thing it had rebooted
and reiserfs said that it couldn't find a file system on the IDE set. Now,
after carefully examined the situation I think that it might have something
to do with LVM. It seems like one disk is missing from the vg.

This is the things I've done (from notes on paper, so it's not the real
output from the screen):

> fdisk /dev/hda

Device	boot	start	end	blocks	id	system
/dev/hda1	*	1	13	104391	03	Linux		
/dev/hda2		14	2482	19832242	5	Extended
/dev/hda3		14	2482	19832211	8e	Linux lvm

> fdisk /dev/hdb
Device	boot	start	end	blocks	id	system
/dev/hdb1	*	1	5005	40202631	8e	Linux lvm

> fdisk /dev/hdc
Device	boot	start	end	blocks	id	system
/dev/hdc1	*	1	19929	160079661	8e	Linux lvm

> vgscan
vgscan -- found active volume group "SystemVolume"
vgscan  -- ERROR "vg_read_with_pv_and_lv() : current PV" can't get data of
volume group "DataVolume" from physical volume(s)
...

> pvscan
-- active PV "/dev/scsi//host0/bus0/target8/lun0/part1" of VG "SystemVolume"
[4.22 GB/0 free]
-- active PV "/dev/scsi//host0/bus0/target9/lun0/part1" of VG "SystemVolume"
[4.22 GB/0 free]
-- active PV "/dev/scsi//host0/bus0/target10/lun0/part1" of VG
"SystemVolume" [4.22 GB/0 free]
-- inactive PV "/dev/ide/host0/bus0/target0/lun0/part5" is associated to
unknown VG "DataVolume" (run vgscan)
-- inactive PV "/dev/ide/host0/bus0/target1/lun0/part1" is associated to
unknown VG "DataVolume" (run vgscan)
-- total : 5 [70.01 GB] / in use : 5 [70.01 GB] / in no VG : 0 [0]


And, as I told you, this is my first try with linux lvm. Not sure what to do
now and I rather try to find some answers before I make a mess on my own. I
would like to save the files from the system if possible, but it's nothing
that I can't replace (what I can't replace I have on tapes also).

Any suggestions how to solve this situation?

Thank's for reading.

/Roger E

Re: [linux-lvm] First try, first fail with LVM.

[Heinz Mauelshagen]

On Wed, Jan 14, 2004 at 10:46:43PM +0100, Roger Eriksson wrote:
> I'm new at this list and new with linux lvm.
> 
> The story this far. I bought a new hard disk to my file server at home and
> decided to reinstall the thing with a newer version of Linux (Red Hat 6.2
> before, now Mandrake 9.2). I decided to go with LVM and reiserfs, during the
> setup of Mandrake 9.2.

Roger,

we've got one problem report which might be related, where LVM fails to
create UUIDs on Mandrake which is probably related to /dev/urandom used
by LVM to retrieve random numbers. Check if /dev/urandom exists and
is readable.

Please check with pvdisplay on all you PVs and "vgdisplay -D" if
they have UUIDs set fine and report results here.

Regards,
Heinz    -- The LVM Guy --


> 
> On the server I have two sets of disks. Three 4.6 Gb SCSI disks that I
> joined together using LVM and three larger IDE disks that I also joined
> together. Everything went well and I started to copy back the files that I
> earlier copied to my Window and Linux desktop computers. No problems with
> that and since it got late a set the computers to move the files back and
> went to bed. The next morning when I checked the damn thing it had rebooted
> and reiserfs said that it couldn't find a file system on the IDE set. Now,
> after carefully examined the situation I think that it might have something
> to do with LVM. It seems like one disk is missing from the vg.
> 
> This is the things I've done (from notes on paper, so it's not the real
> output from the screen):
> 
> > fdisk /dev/hda
> 
> Device	boot	start	end	blocks	id	system
> /dev/hda1	*	1	13	104391	03	Linux		
> /dev/hda2		14	2482	19832242	5	Extended
> /dev/hda3		14	2482	19832211	8e	Linux lvm
> 
> > fdisk /dev/hdb
> Device	boot	start	end	blocks	id	system
> /dev/hdb1	*	1	5005	40202631	8e	Linux lvm
> 
> > fdisk /dev/hdc
> Device	boot	start	end	blocks	id	system
> /dev/hdc1	*	1	19929	160079661	8e	Linux lvm
> 
> > vgscan
> vgscan -- found active volume group "SystemVolume"
> vgscan  -- ERROR "vg_read_with_pv_and_lv() : current PV" can't get data of
> volume group "DataVolume" from physical volume(s)
> ...
> 
> > pvscan
> -- active PV "/dev/scsi//host0/bus0/target8/lun0/part1" of VG "SystemVolume"
> [4.22 GB/0 free]
> -- active PV "/dev/scsi//host0/bus0/target9/lun0/part1" of VG "SystemVolume"
> [4.22 GB/0 free]
> -- active PV "/dev/scsi//host0/bus0/target10/lun0/part1" of VG
> "SystemVolume" [4.22 GB/0 free]
> -- inactive PV "/dev/ide/host0/bus0/target0/lun0/part5" is associated to
> unknown VG "DataVolume" (run vgscan)
> -- inactive PV "/dev/ide/host0/bus0/target1/lun0/part1" is associated to
> unknown VG "DataVolume" (run vgscan)
> -- total : 5 [70.01 GB] / in use : 5 [70.01 GB] / in no VG : 0 [0]
> 
> 
> And, as I told you, this is my first try with linux lvm. Not sure what to do
> now and I rather try to find some answers before I make a mess on my own. I
> would like to save the files from the system if possible, but it's nothing
> that I can't replace (what I can't replace I have on tapes also).
> 
> Any suggestions how to solve this situation?
> 
> Thank's for reading.
> 
> /Roger E
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/

*** Software bugs are stupid.
    Nevertheless it needs not so stupid people to solve them ***

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Heinz Mauelshagen                                 Sistina Software Inc.
Senior Consultant/Developer                       Am Sonnenhang 11
                                                  56242 Marienrachdorf
                                                  Germany
Mauelshagen@Sistina.com                           +49 2626 141200
                                                       FAX 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

RE: [linux-lvm] First try, first fail with LVM.

[Roger Eriksson]

Thank you for answereing.

It turned out to be a problem with the power supply as it couldn't output
enough power to run six farly new disks (3 IDE & 3 SCSI) . Eight old disk
worked just fine (3 + 5) though.

I managed to restore most of the data from the NTFS file systems so I
decided to reinstall the Linux server and run some tests with it and found
out about the power problem. Always troble with the same disk but when I
moved it to another computer and ran Maxtor's hard disk test program several
times and no problems. I will still be a bit suspicious about it though,
until it's proven itself..

Thanks again.