All of lore.kernel.org
 help / color / mirror / Atom feed
* [linux-lvm] simple LVM disk failure recover
@ 2011-08-31  9:26 benry
  2011-09-24 15:35 ` benry
  0 siblings, 1 reply; 4+ messages in thread
From: benry @ 2011-08-31  9:26 UTC (permalink / raw)
  To: linux-lvm

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi,
I'm quite new to LVM and I think to have a quite simple problem, but I
can not find a clear answer to it and, since it is a quite critical
problem, I'm asking to you for the right solution before messing up
everything.

I have an LVM volume made by two raid sda and sdb, with an XFS
filesystem on it; sda had a failure and its data is completely lost...
what I would like to do is to restore at least what is contained in sdb.
(The situation displayed by "lvdisplay -mP" is reported at the end of
this mail).

I tried this command: "vgchange -P -a y raid", but the device
/dev/raid/data does not exist and the /dev/mapper/raid-data I used to
mount the volume seems to be not accessible:

# xfs_check /dev/mapper/raid-data
xfs_check: /dev/mapper/raid-data is invalid (cannot read first 512 bytes)

Is the data contained in sdb recoverable? (as far as I understood, it
should) And if it is the case, which should be the right procedure to
restore it?


Thanks in advance for your help.

Enrico

# lvdisplay -mP
  Partial mode. Incomplete volume groups will be activated read-only.
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  /dev/sda: read failed after 0 of 4096 at 19989989883904: Input/output
error
  /dev/sda: read failed after 0 of 4096 at 19989989941248: Input/output
error
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  /dev/sda: read failed after 0 of 4096 at 4096: Input/output error
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  Couldn't find device with uuid 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'.
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  /dev/sda: read failed after 0 of 4096 at 0: Input/output error
  Couldn't find device with uuid 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'.
  --- Logical volume ---
  LV Name                /dev/raid/data
  VG Name                raid
  LV UUID                4rjFt7-ZuQX-f3n4-iDVR-2gJv-5cFf-77gG5f
  LV Write Access        read/write
  LV Status              available
  # open                 0
  LV Size                36.36 TB
  Current LE             9531968
  Segments               2
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     256
  Block device           254:0

  --- Segments ---
  Logical extent 0 to 4765983:
    Type                linear
    Physical volume     unknown device
    Physical extents    0 to 4765983

  Logical extent 4765984 to 9531967:
    Type                linear
    Physical volume     /dev/sdb1
    Physical extents    0 to 4765983
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)

iEYEARECAAYFAk5d/lEACgkQOideAxdXzI87ZACgll/dlythiZi3REwdJRMwj5Vf
I2kAnir0p+Owr+2Hnanzu2K5Zf1Gg0+k
=Odjy
-----END PGP SIGNATURE-----

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [linux-lvm] simple LVM disk failure recover
  2011-08-31  9:26 [linux-lvm] simple LVM disk failure recover benry
@ 2011-09-24 15:35 ` benry
  2011-09-24 19:52   ` Ray Morris
  0 siblings, 1 reply; 4+ messages in thread
From: benry @ 2011-09-24 15:35 UTC (permalink / raw)
  To: LVM general discussion and development

Hi,
sorry for the top-posting, but in this case it seems a good choice...
the situation evolved from my last post:

The sda raid was lost completely due to an hardware problem of the raid
card, actually there is a new board and a new empty raid... how could I
convince LVM to use it, in place of the old one? (the uuid are
different), but moreover, can I restore in some way the content of the
other raid that is still full of data?

LVM was configured to use sda and sdb in linear mode... not stripping...
so theoretically data in sdb should be readable, right?

Please answer me... also RTFM with a link to the right manual would be
kindly appreciated (if you know that an answer for my question exist in
the manual) ;)


Thanks in advance,

Enrico


On 08/31/2011 11:26 AM, benry wrote:
> Hi,
> I'm quite new to LVM and I think to have a quite simple problem, but I
> can not find a clear answer to it and, since it is a quite critical
> problem, I'm asking to you for the right solution before messing up
> everything.
> 
> I have an LVM volume made by two raid sda and sdb, with an XFS
> filesystem on it; sda had a failure and its data is completely lost...
> what I would like to do is to restore at least what is contained in sdb.
> (The situation displayed by "lvdisplay -mP" is reported at the end of
> this mail).
> 
> I tried this command: "vgchange -P -a y raid", but the device
> /dev/raid/data does not exist and the /dev/mapper/raid-data I used to
> mount the volume seems to be not accessible:
> 
> # xfs_check /dev/mapper/raid-data
> xfs_check: /dev/mapper/raid-data is invalid (cannot read first 512 bytes)
> 
> Is the data contained in sdb recoverable? (as far as I understood, it
> should) And if it is the case, which should be the right procedure to
> restore it?
> 
> 
> Thanks in advance for your help.
> 
> Enrico
> 
> # lvdisplay -mP
>   Partial mode. Incomplete volume groups will be activated read-only.
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   /dev/sda: read failed after 0 of 4096 at 19989989883904: Input/output
> error
>   /dev/sda: read failed after 0 of 4096 at 19989989941248: Input/output
> error
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   /dev/sda: read failed after 0 of 4096 at 4096: Input/output error
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   Couldn't find device with uuid 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'.
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
>   Couldn't find device with uuid 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'.
>   --- Logical volume ---
>   LV Name                /dev/raid/data
>   VG Name                raid
>   LV UUID                4rjFt7-ZuQX-f3n4-iDVR-2gJv-5cFf-77gG5f
>   LV Write Access        read/write
>   LV Status              available
>   # open                 0
>   LV Size                36.36 TB
>   Current LE             9531968
>   Segments               2
>   Allocation             inherit
>   Read ahead sectors     auto
>   - currently set to     256
>   Block device           254:0
> 
>   --- Segments ---
>   Logical extent 0 to 4765983:
>     Type                linear
>     Physical volume     unknown device
>     Physical extents    0 to 4765983
> 
>   Logical extent 4765984 to 9531967:
>     Type                linear
>     Physical volume     /dev/sdb1
>     Physical extents    0 to 4765983

_______________________________________________
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/

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [linux-lvm] simple LVM disk failure recover
  2011-09-24 15:35 ` benry
@ 2011-09-24 19:52   ` Ray Morris
  2011-09-25  1:35     ` benry
  0 siblings, 1 reply; 4+ messages in thread
From: Ray Morris @ 2011-09-24 19:52 UTC (permalink / raw)
  To: LVM general discussion and development; +Cc: benry

.. also RTFM with a link to the right manual would be
> kindly appreciated (if you know that an answer for my question exist
> in the manual) ;)

man lvm | grep -5 partial


       -P | --partial
              When set, the tools will do their best to provide access
       to vol- ume groups that are only partially available (one or
       more physi- cal  volumes  belonging to the volume group are
       missing from the system).  Where part of a logical volume is
       missing,  /dev/ioer- ror  will  be  substituted, and you could
       use dmsetup (8) to set this up to return I/O errors when
       accessed, or create  it  as  a large  block  device of nulls.
       Metadata may not be changed with


You probably do not need:
man pvcreate | grep -2 uuid

       -u, --uuid uuid
              Specify the uuid for the device.  Without this option,
       pvcreate generates a random uuid.  All of your physical volumes
       must have unique uuids.  You need to use this option  before
       restoring  a backup  of LVM metadata onto a replacement device -
       see vgcfgre- store(8).


> LVM was configured to use sda and sdb in linear mode... not
> stripping... so theoretically data in sdb should be readable, right?

The data on sdb is there, but it's probably not a correct filesystem, 
so reading it may be difficult.
-- 
Ray Morris
support@bettercgi.com

Strongbox - The next generation in site security:
http://www.bettercgi.com/strongbox/

Throttlebox - Intelligent Bandwidth Control
http://www.bettercgi.com/throttlebox/

Strongbox / Throttlebox affiliate program:
http://www.bettercgi.com/affiliates/user/register.php




On Sat, 24 Sep 2011 17:35:13 +0200
benry <benry@autistici.org> wrote:

> Hi,
> sorry for the top-posting, but in this case it seems a good choice...
> the situation evolved from my last post:
> 
> The sda raid was lost completely due to an hardware problem of the
> raid card, actually there is a new board and a new empty raid... how
> could I convince LVM to use it, in place of the old one? (the uuid are
> different), but moreover, can I restore in some way the content of the
> other raid that is still full of data?
> 
> LVM was configured to use sda and sdb in linear mode... not
> stripping... so theoretically data in sdb should be readable, right?
> 
> Please answer me... also RTFM with a link to the right manual would be
> kindly appreciated (if you know that an answer for my question exist
> in the manual) ;)
> 
> 
> Thanks in advance,
> 
> Enrico
> 
> 
> On 08/31/2011 11:26 AM, benry wrote:
> > Hi,
> > I'm quite new to LVM and I think to have a quite simple problem,
> > but I can not find a clear answer to it and, since it is a quite
> > critical problem, I'm asking to you for the right solution before
> > messing up everything.
> > 
> > I have an LVM volume made by two raid sda and sdb, with an XFS
> > filesystem on it; sda had a failure and its data is completely
> > lost... what I would like to do is to restore at least what is
> > contained in sdb. (The situation displayed by "lvdisplay -mP" is
> > reported at the end of this mail).
> > 
> > I tried this command: "vgchange -P -a y raid", but the device
> > /dev/raid/data does not exist and the /dev/mapper/raid-data I used
> > to mount the volume seems to be not accessible:
> > 
> > # xfs_check /dev/mapper/raid-data
> > xfs_check: /dev/mapper/raid-data is invalid (cannot read first 512
> > bytes)
> > 
> > Is the data contained in sdb recoverable? (as far as I understood,
> > it should) And if it is the case, which should be the right
> > procedure to restore it?
> > 
> > 
> > Thanks in advance for your help.
> > 
> > Enrico
> > 
> > # lvdisplay -mP
> >   Partial mode. Incomplete volume groups will be activated
> > read-only. /dev/sda: read failed after 0 of 4096 at 0: Input/output
> > error /dev/sda: read failed after 0 of 4096 at 0: Input/output error
> >   /dev/sda: read failed after 0 of 4096 at 19989989883904:
> > Input/output error
> >   /dev/sda: read failed after 0 of 4096 at 19989989941248:
> > Input/output error
> >   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
> >   /dev/sda: read failed after 0 of 4096 at 4096: Input/output error
> >   /dev/sda: read failed after 0 of 4096 at 0: Input/output error
> >   Couldn't find device with uuid
> > 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'. /dev/sda: read failed
> > after 0 of 4096 at 0: Input/output error /dev/sda: read failed
> > after 0 of 4096 at 0: Input/output error Couldn't find device with
> > uuid 'AkLdC5-wPgi-hDbv-Sqf3-2wBv-hBzu-2OhDaQ'. --- Logical volume
> > --- LV Name                /dev/raid/data
> >   VG Name                raid
> >   LV UUID                4rjFt7-ZuQX-f3n4-iDVR-2gJv-5cFf-77gG5f
> >   LV Write Access        read/write
> >   LV Status              available
> >   # open                 0
> >   LV Size                36.36 TB
> >   Current LE             9531968
> >   Segments               2
> >   Allocation             inherit
> >   Read ahead sectors     auto
> >   - currently set to     256
> >   Block device           254:0
> > 
> >   --- Segments ---
> >   Logical extent 0 to 4765983:
> >     Type                linear
> >     Physical volume     unknown device
> >     Physical extents    0 to 4765983
> > 
> >   Logical extent 4765984 to 9531967:
> >     Type                linear
> >     Physical volume     /dev/sdb1
> >     Physical extents    0 to 4765983
> 
> _______________________________________________
> 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/
> 
> _______________________________________________
> 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/
> 

^ permalink raw reply	[flat|nested] 4+ messages in thread
* Re: [linux-lvm] simple LVM disk failure recover
  2011-09-24 19:52   ` Ray Morris
@ 2011-09-25  1:35     ` benry
  0 siblings, 0 replies; 4+ messages in thread
From: benry @ 2011-09-25  1:35 UTC (permalink / raw)
  To: LVM general discussion and development; +Cc: support

On 09/24/2011 09:52 PM, Ray Morris wrote:
> .. also RTFM with a link to the right manual would be
>> kindly appreciated (if you know that an answer for my question exist
>> in the manual) ;)
> 
> man lvm | grep -5 partial

Ok, thanks very much for the answer, but I already tried the "partial"
option:

On 08/31/2011 11:26 AM, benry wrote:
> > I tried this command: "vgchange -P -a y raid", but the device
> > /dev/raid/data does not exist anyway.

Maybe I do not understand very well the part where it says "/dev/ioerror
will  be substituted, and you could use dmsetup (8) to set this up to
return I/O errors when accessed, or create  it  as  a large  block
device of nulls."
Is this optional or I have to do something with dmsetup to let the
command "vgchange -P -a y raid" work correctly?

Maybe I am missing what is the right purpose of this /dev/ioerror device.

> You probably do not need:
> man pvcreate | grep -2 uuid

Ok, so, this is needed to let the new pv (sda RAID) to be initialized
with the same uuid of the old one... right?
If I use it on the new drive, LVM will believe to have the old pv and
will make the volume available, apart from the fact that it will not
find data in it.
This makes sense to me ;)

>> LVM was configured to use sda and sdb in linear mode... not
>> stripping... so theoretically data in sdb should be readable, right?
> 
> The data on sdb is there, but it's probably not a correct filesystem, 
> so reading it may be difficult.

Ok, this makes sense too... what I'll try to do (since the lv was
containing an XFS filesystem) is to use the xfs_repair command, that as
far as I understood should look for pieces of an XFS filesystem and try
to repair it.

In conclusion, what I could try to do is:
1) create a new pv in the new RAID with the same uuid of the old one
(pvcreate)
2) activate the lv (vgchange)
3) try xfs_repair on the lv crossing my fingers

Right?


Thanks very much for the help... I read a lot of documentation, but
since I do not have a lot of experience with LVM and I am really afraid
of damaging also the remaining data, I kindly appreciate to have some of
your comments.

Enrico

^ permalink raw reply	[flat|nested] 4+ messages in thread
end of thread, other threads:[~2011-09-25  1:35 UTC | newest]

Thread overview: 4+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2011-08-31  9:26 [linux-lvm] simple LVM disk failure recover benry
2011-09-24 15:35 ` benry
2011-09-24 19:52   ` Ray Morris
2011-09-25  1:35     ` benry

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.