[linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[d.levitin]

Hello,
I have a problem with my vg00 - the LVM doesn't have it any more. I found
such postings in the list archive, but I don't want to restore - as I think.
To the problem: 
I had 10 disks in my Computer (/dev/sda.../dev/sdj) in the vg00. Then I have
2 of them (/dev/sdj and /dev/sdi) unplugged. I hoped it would be no problem
for the LVM to work with the rest 8, but it doesn't do. I dont' worry about
the volumes on the sdi and sdj, but the other volumes are a problem...
So why the LVM dosn't see the other disks as well?

Thanks a lot for your help
Dmitri
 

# vgscan
vgscan -- reading all physical volumes (this may take a while...)
vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data
of volume group "vg00" from physical volume(s)
vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
vgscan -- WARNING: This program does not do a VGDA backup of your
volume group

# pvscan
pvscan -- reading all physical volumes (this may take a while...)
pvscan -- inactive PV "/dev/sda4"  is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdb3"  is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdc"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdd"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sde"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdf"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdg"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- inactive PV "/dev/sdh"   is associated to unknown VG "vg00"
(run vgscan)
pvscan -- total: 8 [200.21 GB] / in use: 8 [200.21 GB] / in no VG: 0
[0]


-- 
GMX - Die Kommunikationsplattform im Internet.
http://www.gmx.net

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[James Hawtin]

On Sat, 3 Aug 2002 d.levitin@gmx.de wrote:

> Hello,
> I have a problem with my vg00 - the LVM doesn't have it any more. I found
> such postings in the list archive, but I don't want to restore - as I think.
> To the problem:
> I had 10 disks in my Computer (/dev/sda.../dev/sdj) in the vg00. Then I have
> 2 of them (/dev/sdj and /dev/sdi) unplugged. I hoped it would be no problem
> for the LVM to work with the rest 8, but it doesn't do. I dont' worry about
> the volumes on the sdi and sdj, but the other volumes are a problem...
> So why the LVM dosn't see the other disks as well?
>
> Thanks a lot for your help
> Dmitri
>

Because LVM does not currently mount part volume groups for safety...
I believe at some in in the future this might be solved.

Do you still have /dev/sdj and /dev/sdi you say they are unpluged at the
moment and you don't want access to the info on them... if they have data
that is split onto other pvs use pvmove move so only have lvs you don't
what on those two pvs then use vgsplit to remove them from the vg, then
you can unplug those two disks as they will be in a new volume group.

lv = logical volume
pv = physical volume
vg = volume group


james

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[d.levitin]

> 
> 
> On Sat, 3 Aug 2002 d.levitin@gmx.de wrote:
> 
> > Hello,
> > I have a problem with my vg00 - the LVM doesn't have it any more. I
> found
> > such postings in the list archive, but I don't want to restore - as I
> think.
> > To the problem:
> > I had 10 disks in my Computer (/dev/sda.../dev/sdj) in the vg00. Then I
> have
> > 2 of them (/dev/sdj and /dev/sdi) unplugged. I hoped it would be no
> problem
> > for the LVM to work with the rest 8, but it doesn't do. I dont' worry
> about
> > the volumes on the sdi and sdj, but the other volumes are a problem...
> > So why the LVM dosn't see the other disks as well?
> >
> > Thanks a lot for your help
> > Dmitri
> >
> 
> Because LVM does not currently mount part volume groups for safety...
> I believe at some in in the future this might be solved.
> 
> Do you still have /dev/sdj and /dev/sdi you say they are unpluged at the
> moment and you don't want access to the info on them... if they have data
> that is split onto other pvs use pvmove move so only have lvs you don't
> what on those two pvs then use vgsplit to remove them from the vg, then
> you can unplug those two disks as they will be in a new volume group.
> 
> lv = logical volume
> pv = physical volume
> vg = volume group
> 
> 
> james
> 
> 

The 2 disks are for us temporarily unavailable (problem with a controller)
and I can't wait untill I can use them (so no pvmove :-(  ). Can I tell LVM it
have only with the other 8 disks to go on ?

Dmitri

-- 
GMX - Die Kommunikationsplattform im Internet.
http://www.gmx.net

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[sander]

(Response to message from d.levitin@gmx.de, d.levitin@gmx.de on Saturday, August 03, 2002 2:05:30 PM)

Hello,

Couldn't that have possibly disastrous results, since files or
directories may have their PE's mapped partly on an available disk and
partly on one that's missing?

-- 

Regards,
sander

Saturday, August 03, 2002, 2:05:30 PM, you wrote:


> The 2 disks are for us temporarily unavailable (problem with a controller)
> and I can't wait untill I can use them (so no pvmove :-(  ). Can I tell LVM it
> have only with the other 8 disks to go on ?

> Dmitri

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[d.levitin]

We should simply take an example: 
a VG vg00 resides on 10 PVs (sda...sdj) and has 4 LVs in it: lvol1...lvol4.
Only lvol4 resides on sdaj and sdi, and they become unavailable. For the
other LVs in the vg00 are all PVs still OK, so I should be able to use them, such
as HP-UX or Veritas Volume Manages do. Indeed I have now a situation, that
the entire vg00 ist not available!

Help! (F1! F1!  ;-))  )

Dmitri

> (Response to message from d.levitin@gmx.de, d.levitin@gmx.de on Saturday,
> August 03, 2002 2:05:30 PM)
> 
> Hello,
> 
> Couldn't that have possibly disastrous results, since files or
> directories may have their PE's mapped partly on an available disk and
> partly on one that's missing?
> 
> -- 
> 
> Regards,
> sander
> 
> Saturday, August 03, 2002, 2:05:30 PM, you wrote:
> 
> 
> > The 2 disks are for us temporarily unavailable (problem with a
> controller)
> > and I can't wait untill I can use them (so no pvmove :-(  ). Can I tell
> LVM it
> > have only with the other 8 disks to go on ?
> 
> > Dmitri
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://www.sistina.com/lvm/Pages/howto.html
> 

-- 
GMX - Die Kommunikationsplattform im Internet.
http://www.gmx.net

[linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[`VL]

[-- Attachment #1: Type: text/plain, Size: 1529 bytes --]

Dear ALL, <linux-lvm@redhat.com>

I have a problem with LVM volume(details below) probably after bad shutdown:

There are 2 volumes: lvm2 and lvm1 (just names, type is lvm-1). lvm2 is fine
and mounted,
and lvm1 has problems:

# pvscan
pvscan -- reading all physical volumes (this may take a while...)
pvscan -- ACTIVE   PV "/dev/sda1" of VG "lvm2" [372.56 GB / 0 free]
pvscan -- ACTIVE   PV "/dev/sdb1" of VG "lvm2" [372.56 GB / 0 free]
pvscan -- inactive PV "/dev/hdg1"  is associated to unknown VG "lvm1" (run
vgscan)
pvscan -- inactive PV "/dev/hdf1"  is associated to unknown VG "lvm1" (run
vgscan)
pvscan -- inactive PV "/dev/hdb1"  is associated to unknown VG "lvm1" (run
vgscan)
pvscan -- total: 5 [1.36 TB] / in use: 5 [1.36 TB] / in no VG: 0 [0]

# vgscan
vgscan -- reading all physical volumes (this may take a while...)
vgscan -- found active volume group "lvm2"
vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of
volume group "lvm1" from physical volume(s)
vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
vgscan -- WARNING: This program does not do a VGDA backup of your volume
groups

 ls -l /dev/lvm*
crw-r-----  1 root root 109, 0 2005-03-27 17:11 /dev/lvm

/dev/lvm1:
total 0
crw-r-----  1 root disk 109, 0 2007-01-05 17:14 group

/dev/lvm2:
total 0
crw-r-----  1 root disk 109, 1 2007-01-09 21:22 group
brw-rw----  1 root disk  58, 1 2007-01-09 21:22 lv2


System description:

kernel 2.4.31 vanilla
slackware-10.2.0
lvm-1.0.8-i486-1

any help will be greatly appreciated.

[-- Attachment #2: Type: text/html, Size: 1969 bytes --]

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[Heinz Mauelshagen]

On Tue, Jan 09, 2007 at 09:26:03PM +0300, `VL wrote:
> Dear ALL, <linux-lvm@redhat.com>
> 
> I have a problem with LVM volume(details below) probably after bad shutdown:

Do you assume that, because your system died during
a running configuration change on "lvm1" ?

Are all PVs of "lvm1" accessible ?

If so, you need to fetch your most recent valid VG metadata archive
(see 'man vgcfgrestore'; option '-ll' to look at archive content)
and restore it. 'vgscan ; vgchange -ay' afterwards.

Regards,
Heinz    -- The LVM Guy --


> 
> There are 2 volumes: lvm2 and lvm1 (just names, type is lvm-1). lvm2 is fine
> and mounted,
> and lvm1 has problems:
> 
> # pvscan
> pvscan -- reading all physical volumes (this may take a while...)
> pvscan -- ACTIVE   PV "/dev/sda1" of VG "lvm2" [372.56 GB / 0 free]
> pvscan -- ACTIVE   PV "/dev/sdb1" of VG "lvm2" [372.56 GB / 0 free]
> pvscan -- inactive PV "/dev/hdg1"  is associated to unknown VG "lvm1" (run
> vgscan)
> pvscan -- inactive PV "/dev/hdf1"  is associated to unknown VG "lvm1" (run
> vgscan)
> pvscan -- inactive PV "/dev/hdb1"  is associated to unknown VG "lvm1" (run
> vgscan)
> pvscan -- total: 5 [1.36 TB] / in use: 5 [1.36 TB] / in no VG: 0 [0]
> 
> # vgscan
> vgscan -- reading all physical volumes (this may take a while...)
> vgscan -- found active volume group "lvm2"
> vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of
> volume group "lvm1" from physical volume(s)
> vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
> vgscan -- WARNING: This program does not do a VGDA backup of your volume
> groups
> 
> ls -l /dev/lvm*
> crw-r-----  1 root root 109, 0 2005-03-27 17:11 /dev/lvm
> 
> /dev/lvm1:
> total 0
> crw-r-----  1 root disk 109, 0 2007-01-05 17:14 group
> 
> /dev/lvm2:
> total 0
> crw-r-----  1 root disk 109, 1 2007-01-09 21:22 group
> brw-rw----  1 root disk  58, 1 2007-01-09 21:22 lv2
> 
> 
> System description:
> 
> kernel 2.4.31 vanilla
> slackware-10.2.0
> lvm-1.0.8-i486-1
> 
> any help will be greatly appreciated.

> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.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                                 Red Hat GmbH
Consulting Development Engineer                   Am Sonnenhang 11
Storage Development                               56242 Marienrachdorf
                                                  Germany
Mauelshagen@RedHat.com                            PHONE +49  171 7803392
                                                  FAX   +49 2626 924446
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[`VL]

[-- Attachment #1: Type: text/plain, Size: 1018 bytes --]

>>Do you assume that, because your system died during
>>a running configuration change on "lvm1" ?
I really don`t know. The system is a lan server which has no particular
owner and good maintance. possibly ;-)

>>Are all PVs of "lvm1" accessible ?
i was happy to find a backup of logical volume and it says that there were 1
more harddrive, which
is now NOT accessible. so i assume that this is the problem, since vgscan
complains about
'current PV not accessible'. I`ll try to do my best in searching for this
hard drive, but chances i`ll find
it safe are really low (since anyone who took it, really needed it for
something =)

Now i have unaccessible lvm1 volume, 2 hard drives (hdf and hdg) and backup
file that contains
information about lvm1 when it had 3 hard drives(+hdh).

Are there are any ways to get data from remaining drives? How to remove
information about 3rd drive from volume?
Is it possible?

P.S. sorry that i didn`t understand immediately that problem was because of
loosing 1 of drives from lvm.

[-- Attachment #2: Type: text/html, Size: 1120 bytes --]

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[Joel Rees]

Theis looks a little like something I just went through.

Check

     http://tldp.org/HOWTO/LVM-HOWTO/  for general stuff.

also

     man pvcreate
     man vgcfgrestore

and if you want to look at my descriptions of what I did,

      https://www.redhat.com/archives/linux-lvm/2007-January/ 
msg00003.html
      https://www.redhat.com/archives/fedora-list/2007-January/ 
msg00770.html

I can't say whether that will be more confusion than help. Good luck.

On 平成 19/01/10, at 3:26, `VL wrote:

> Dear ALL,
>
> I have a problem with LVM volume(details below) probably after bad  
> shutdown:
>
> There are 2 volumes: lvm2 and lvm1 (just names, type is lvm-1).  
> lvm2 is fine and mounted,
> and lvm1 has problems:
>
> # pvscan
> pvscan -- reading all physical volumes (this may take a while...)
> pvscan -- ACTIVE   PV "/dev/sda1" of VG "lvm2" [372.56 GB / 0 free]
> pvscan -- ACTIVE   PV "/dev/sdb1" of VG "lvm2" [ 372.56 GB / 0 free]
> pvscan -- inactive PV "/dev/hdg1"  is associated to unknown VG  
> "lvm1" (run vgscan)
> pvscan -- inactive PV "/dev/hdf1"  is associated to unknown VG  
> "lvm1" (run vgscan)
> pvscan -- inactive PV "/dev/hdb1"  is associated to unknown VG  
> "lvm1" (run vgscan)
> pvscan -- total: 5 [1.36 TB] / in use: 5 [1.36 TB] / in no VG: 0 [0]
>
> # vgscan
> vgscan -- reading all physical volumes (this may take a while...)
> vgscan -- found active volume group "lvm2"
> vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get  
> data of volume group "lvm1" from physical volume(s)
> vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
> vgscan -- WARNING: This program does not do a VGDA backup of your  
> volume groups
>
>  ls -l /dev/lvm*
> crw-r-----  1 root root 109, 0 2005-03-27 17:11 /dev/lvm
>
> /dev/lvm1:
> total 0
> crw-r-----  1 root disk 109, 0 2007-01-05 17:14 group
>
> /dev/lvm2:
> total 0
> crw-r-----  1 root disk 109, 1 2007-01-09 21:22 group
> brw-rw----  1 root disk  58, 1 2007-01-09 21:22 lv2
>
>
> System description:
>
> kernel 2.4.31 vanilla
> slackware-10.2.0
> lvm-1.0.8-i486-1
>
> any help will be greatly appreciated.
>
>
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@redhat.com
> https://www.redhat.com/mailman/listinfo/linux-lvm
> read the LVM HOW-TO at http://tldp.org/HOWTO/LVM-HOWTO/

Re: [linux-lvm] vgscan -- ERROR "vg_read_with_pv_and_lv(): current PV" can't get data of volume group

[`VL]

[-- Attachment #1: Type: text/plain, Size: 1132 bytes --]

Thanks for everyone!

I`ve got some ideas that helped me to get data back. Here is summary, that
someone maybe find usefull:

volume group lost one of it`s physical volumes and refused to mount.
So, i had to create new physical volume with same metadata (UUID and others)
as old one,
Metadata was saved in backup.The only problem i had that new harddisk i had
was not hdh (as one those
at volume group was), but hdb. Thus, vgcfgrestore refused to work, telling
that there are no such drive
in volume group. So, i edited manually binary backup file of lvm and changed
just one letter in the name of disk.
it worked. Also,i had to set same UID for new physical volume as it was for
old drive. I achieved this by
installing lvm2 tools and using pvcreate from there. After all those
manipulations, vgscan reported about
inactive volume group, which was successfully activated.

P.S. The most funny part of all this story is that there was 23 megs of data
on that group. Person, who
asked me to help with maintance, remembered that there was some sort of
'cleaning' on that server. =)

P.P.S. Thanks for your job, LVM people! LVM rules.

[-- Attachment #2: Type: text/html, Size: 1197 bytes --]