All of lore.kernel.org
 help / color / mirror / Atom feed
* [linux-lvm] removing a failed disk from VG
@ 2003-01-27 14:29 Alois Schneider
  2003-01-28  4:50 ` Heinz J . Mauelshagen
  0 siblings, 1 reply; 4+ messages in thread
From: Alois Schneider @ 2003-01-27 14:29 UTC (permalink / raw)
  To: linux-lvm

I had a VG scsi_vg consisting of 4 PV's (sda1, sdb1, hdb1,hdc1). 
Now /dev/sdb1 failed and was no longer recogniced by th SCSI-
controller. I had to remove it.

vgscan now fails with the following error:

vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data 
of volume group "scsi_vg" from physical volume(s)
vgscan -- reading all physical volumes (this may take a while...)
vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
vgscan -- WARNING: This program does not do a VGDA backup of your 
volume group
 
vgchange -ay gives
-- no volume group found

The remaining PV's seem to be ok and the failed disk was only 4Gb 
large.

Here is the output of pvdisplay:

--- Physical volume ---
PV Name               /dev/sda1
VG Name               scsi_vg
PV Size               4.04 GB [8466192 secs] / NOT usable 1.88 MB 
[LVM: 125 KB]
PV#                   1
PV Status             available
Allocatable           yes (but full)
Cur LV                1
PE Size (KByte)       4096
Total PE              1033
Free PE               0
Allocated PE          1033
PV UUID               pJsMFv-ci9Q-orNK-3cRr-2aTl-w1N9-TEHDI1


--- Physical volume ---
PV Name               /dev/hdb1
VG Name               scsi_vg
PV Size               76.69 GB [160826652 secs] / NOT usable 4.25 MB 
[LVM: 200 KB]
PV#                   3
PV Status             available
Allocatable           yes (but full)
Cur LV                2
PE Size (KByte)       4096
Total PE              19631
Free PE               0
Allocated PE          19631
PV UUID               u5I40y-Op34-nnVV-aInj-Jekc-VJT7-s2q8OK


--- Physical volume ---
PV Name               /dev/hdc1
VG Name               scsi_vg
PV Size               19.01 GB [39873267 secs] / NOT usable 4.18 MB 
[LVM: 143 KB]
PV#                   5
PV Status             available
Allocatable           yes
Cur LV                1
PE Size (KByte)       4096
Total PE              4866
Free PE               2040
Allocated PE          2826
PV UUID               3PCGsY-hUr6-QEUt-MkoJ-fKlX-24vG-pS0HJY

What can I do? How can I remove the no longer existing PV sdb1 from 
scsi_vg?

Any help will be appreciated.
Regards
Alois

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [linux-lvm] removing a failed disk from VG
  2003-01-27 14:29 [linux-lvm] removing a failed disk from VG Alois Schneider
@ 2003-01-28  4:50 ` Heinz J . Mauelshagen
  2003-01-28  5:08   ` AW: " Kai Iskratsch
  2003-01-28  6:53   ` Alasdair G Kergon
  0 siblings, 2 replies; 4+ messages in thread
From: Heinz J . Mauelshagen @ 2003-01-28  4:50 UTC (permalink / raw)
  To: linux-lvm

On Mon, Jan 27, 2003 at 09:28:08PM +0100, Alois Schneider wrote:
> I had a VG scsi_vg consisting of 4 PV's (sda1, sdb1, hdb1,hdc1). 
> Now /dev/sdb1 failed and was no longer recogniced by th SCSI-
> controller. I had to remove it.
> 

Alois,

then you probably lost data with the gone disk. Backup time :-(
Please make sure that you've got backups of /etc/lvmconf/* and /etc/lvmtab*
before you proceed.

There's 2 options:

a. you get a replacement drive (which can be a loop device)
   of the same size (4G), run pvcreate on it, restore the metadata onto
   it (vgcfgrestore) and run "vgscan ; vgchange -ay"

b. you implement LVM2/device-mapper and run "vgscan ; vgchange -P -ay", which
   will activate your VG without the missing drive. This solution enables
   you to retrieve the data still accessable _but_ won't let you change
   your volume group configuration. We are working on an enhancement supporting
   such changes.

Regards,
Heinz    -- The LVM Guy --


> vgscan now fails with the following error:
> 
> vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data 
> of volume group "scsi_vg" from physical volume(s)
> vgscan -- reading all physical volumes (this may take a while...)
> vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
> vgscan -- WARNING: This program does not do a VGDA backup of your 
> volume group
>  
> vgchange -ay gives
> -- no volume group found
> 
> The remaining PV's seem to be ok and the failed disk was only 4Gb 
> large.
> 
> Here is the output of pvdisplay:
> 
> --- Physical volume ---
> PV Name               /dev/sda1
> VG Name               scsi_vg
> PV Size               4.04 GB [8466192 secs] / NOT usable 1.88 MB 
> [LVM: 125 KB]
> PV#                   1
> PV Status             available
> Allocatable           yes (but full)
> Cur LV                1
> PE Size (KByte)       4096
> Total PE              1033
> Free PE               0
> Allocated PE          1033
> PV UUID               pJsMFv-ci9Q-orNK-3cRr-2aTl-w1N9-TEHDI1
> 
> 
> --- Physical volume ---
> PV Name               /dev/hdb1
> VG Name               scsi_vg
> PV Size               76.69 GB [160826652 secs] / NOT usable 4.25 MB 
> [LVM: 200 KB]
> PV#                   3
> PV Status             available
> Allocatable           yes (but full)
> Cur LV                2
> PE Size (KByte)       4096
> Total PE              19631
> Free PE               0
> Allocated PE          19631
> PV UUID               u5I40y-Op34-nnVV-aInj-Jekc-VJT7-s2q8OK
> 
> 
> --- Physical volume ---
> PV Name               /dev/hdc1
> VG Name               scsi_vg
> PV Size               19.01 GB [39873267 secs] / NOT usable 4.18 MB 
> [LVM: 143 KB]
> PV#                   5
> PV Status             available
> Allocatable           yes
> Cur LV                1
> PE Size (KByte)       4096
> Total PE              4866
> Free PE               2040
> Allocated PE          2826
> PV UUID               3PCGsY-hUr6-QEUt-MkoJ-fKlX-24vG-pS0HJY
> 
> What can I do? How can I remove the no longer existing PV sdb1 from 
> scsi_vg?
> 
> Any help will be appreciated.
> Regards
> Alois
> 
> 
> _______________________________________________
> 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
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

^ permalink raw reply	[flat|nested] 4+ messages in thread

* AW: [linux-lvm] removing a failed disk from VG
  2003-01-28  4:50 ` Heinz J . Mauelshagen
@ 2003-01-28  5:08   ` Kai Iskratsch
  2003-01-28  6:53   ` Alasdair G Kergon
  1 sibling, 0 replies; 4+ messages in thread
From: Kai Iskratsch @ 2003-01-28  5:08 UTC (permalink / raw)
  To: linux-lvm

[-- Attachment #1: Type: text/plain, Size: 5170 bytes --]

would the first method of this also work for me.
I should get an identical harddisk from the vendor since i have still
warranty on it.

I tried allready the second one, and I was lucky to succsessfully get the
data out of 2 from my 3 LVs since they didn't use the broken disk.
The last one which used the broken Harddisk will not mount without a fs
scan, since this is not an option with a readonly FS I still couldn't even
try to get any data from it.

regards

Kai


> -----Ursprungliche Nachricht-----
> Von: linux-lvm-admin@sistina.com [mailto:linux-lvm-admin@sistina.com]Im
> Auftrag von Heinz J . Mauelshagen
> Gesendet: Dienstag, 28. Janner 2003 11:47
> An: linux-lvm@sistina.com
> Betreff: Re: [linux-lvm] removing a failed disk from VG
>
>
> On Mon, Jan 27, 2003 at 09:28:08PM +0100, Alois Schneider wrote:
> > I had a VG scsi_vg consisting of 4 PV's (sda1, sdb1, hdb1,hdc1).
> > Now /dev/sdb1 failed and was no longer recogniced by th SCSI-
> > controller. I had to remove it.
> >
>
> Alois,
>
> then you probably lost data with the gone disk. Backup time :-(
> Please make sure that you've got backups of /etc/lvmconf/* and
> /etc/lvmtab*
> before you proceed.
>
> There's 2 options:
>
> a. you get a replacement drive (which can be a loop device)
>    of the same size (4G), run pvcreate on it, restore the metadata onto
>    it (vgcfgrestore) and run "vgscan ; vgchange -ay"
>
> b. you implement LVM2/device-mapper and run "vgscan ; vgchange -P
> -ay", which
>    will activate your VG without the missing drive. This solution enables
>    you to retrieve the data still accessable _but_ won't let you change
>    your volume group configuration. We are working on an
> enhancement supporting
>    such changes.
>
> Regards,
> Heinz    -- The LVM Guy --
>
>
> > vgscan now fails with the following error:
> >
> > vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data
> > of volume group "scsi_vg" from physical volume(s)
> > vgscan -- reading all physical volumes (this may take a while...)
> > vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
> > vgscan -- WARNING: This program does not do a VGDA backup of your
> > volume group
> >
> > vgchange -ay gives
> > -- no volume group found
> >
> > The remaining PV's seem to be ok and the failed disk was only 4Gb
> > large.
> >
> > Here is the output of pvdisplay:
> >
> > --- Physical volume ---
> > PV Name               /dev/sda1
> > VG Name               scsi_vg
> > PV Size               4.04 GB [8466192 secs] / NOT usable 1.88 MB
> > [LVM: 125 KB]
> > PV#                   1
> > PV Status             available
> > Allocatable           yes (but full)
> > Cur LV                1
> > PE Size (KByte)       4096
> > Total PE              1033
> > Free PE               0
> > Allocated PE          1033
> > PV UUID               pJsMFv-ci9Q-orNK-3cRr-2aTl-w1N9-TEHDI1
> >
> >
> > --- Physical volume ---
> > PV Name               /dev/hdb1
> > VG Name               scsi_vg
> > PV Size               76.69 GB [160826652 secs] / NOT usable 4.25 MB
> > [LVM: 200 KB]
> > PV#                   3
> > PV Status             available
> > Allocatable           yes (but full)
> > Cur LV                2
> > PE Size (KByte)       4096
> > Total PE              19631
> > Free PE               0
> > Allocated PE          19631
> > PV UUID               u5I40y-Op34-nnVV-aInj-Jekc-VJT7-s2q8OK
> >
> >
> > --- Physical volume ---
> > PV Name               /dev/hdc1
> > VG Name               scsi_vg
> > PV Size               19.01 GB [39873267 secs] / NOT usable 4.18 MB
> > [LVM: 143 KB]
> > PV#                   5
> > PV Status             available
> > Allocatable           yes
> > Cur LV                1
> > PE Size (KByte)       4096
> > Total PE              4866
> > Free PE               2040
> > Allocated PE          2826
> > PV UUID               3PCGsY-hUr6-QEUt-MkoJ-fKlX-24vG-pS0HJY
> >
> > What can I do? How can I remove the no longer existing PV sdb1 from
> > scsi_vg?
> >
> > Any help will be appreciated.
> > Regards
> > Alois
> >
> >
> > _______________________________________________
> > 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
> =-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
> =-=-=-=-=-=-
>
> _______________________________________________
> 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/
>

[-- Attachment #2: Type: message/rfc822, Size: 2974 bytes --]

From: "Heinz J . Mauelshagen" <mauelshagen@sistina.com>
To: <linux-lvm@sistina.com>
Subject: Re: [linux-lvm] Broken Harddisk in LVM
Date: Wed, 22 Jan 2003 12:14:46 +0100
Message-ID: <20030122121446.A25896@sistina.com>


Install device-mapper + LVM2 tools _additionally_ in a seperate
location and run "vgscan ; vgchange -a y -P".

That might still leave you with unusable LVs in case their beginning
is gone with the failed disk.

If you decide to do that get back to us in case you've got installation
related questions.

Regards,
Heinz    -- The LVM Guy --

On Tue, Jan 21, 2003 at 11:29:08PM +0100, Kai Iskratsch wrote:
> One hard disk of my lvm System is completely broken, so that even the
> controller cannot find it.
> Is there a possibility to get the lvm running again without the data of
> this disk.
>
> Kai Iskratsch
>
>
>
> _______________________________________________
> 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
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
=-

_______________________________________________
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/

^ permalink raw reply	[flat|nested] 4+ messages in thread

* Re: [linux-lvm] removing a failed disk from VG
  2003-01-28  4:50 ` Heinz J . Mauelshagen
  2003-01-28  5:08   ` AW: " Kai Iskratsch
@ 2003-01-28  6:53   ` Alasdair G Kergon
  1 sibling, 0 replies; 4+ messages in thread
From: Alasdair G Kergon @ 2003-01-28  6:53 UTC (permalink / raw)
  To: linux-lvm

On Tue, Jan 28, 2003 at 11:47:24AM +0100, Heinz J . Mauelshagen wrote:
> b. you implement LVM2/device-mapper and run "vgscan ; vgchange -P -ay", which
>    will activate your VG without the missing drive. This solution enables
>    you to retrieve the data still accessable _but_ won't let you change
>    your volume group configuration. We are working on an enhancement supporting
>    such changes.

So far, 'vgreduce --removemissing' is in CVS; this removes everything
necessary to make a VG consistent again after losing disk(s).
If *any* part of an LV is missing, the *entire* LV gets deleted, so
you should ensure you've made a copy of anything you want to retain
first.  Running first with '--test' shows you which LVs would get
deleted.

Alternatively, if you have replacement disk(s), you can use pvcreate &
vgcfgrestore to put it into the VG in place of the one(s) lost.

Or if you want to do anything differently, you can edit the metadata
in text form and use vgcfgrestore to effect your changes.

Alasdair
-- 
agk@uk.sistina.com

^ permalink raw reply	[flat|nested] 4+ messages in thread

end of thread, other threads:[~2003-01-28  6:53 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2003-01-27 14:29 [linux-lvm] removing a failed disk from VG Alois Schneider
2003-01-28  4:50 ` Heinz J . Mauelshagen
2003-01-28  5:08   ` AW: " Kai Iskratsch
2003-01-28  6:53   ` Alasdair G Kergon

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.