[linux-lvm] i2o devices and lvm Error on LE versus PE

[linux-lvm] i2o devices and lvm Error on LE versus PE

[Gianmichele Murano]

HI  ALL

as first thing, i say hallo to you.

now i tell you my problem:
on a controler raid  (i2o devices) i have a debian distro.
Recently i have enlarged the logical volumes  on this controler.
All seemed ok,  i was able to remount the file systems and to copy and install 
programs and data on resized volumes.

So, i was out of office and the development server was tuned off by person who 
wasn't me (so i don't know what is the matter, if happened any thing ).

later, when in was in office i have turn on the server and the vgscan failed, 
it said " ..omissis... only  found 499 of 500  LE for LV 
/dev/volume1/lgv1(0)" .
I don't undestand what was happened.
It is possible that a bad tuned off make a problem as this?
Any body know the answer?

To solve the trouble i thinked to do in this way:
to reduce the LV lgv1 of some LE , in this way the info about the number of LE  
will rewrite and (i hope) the problem vanish.The idea seems good, isn't?
But i can't do this because the volume group is inactive.
Anybody know a method to active a volume group also if vgscan fail?

Please, i MUST to recove the data on the lv, we don't have a backup policy 
>:^> 

Re: [linux-lvm] i2o devices and lvm Error on LE versus PE

[Heinz J . Mauelshagen]

On Sat, Sep 20, 2003 at 01:54:18PM +0200, Gianmichele Murano wrote:
> HI  ALL
> 
> as first thing, i say hallo to you.
> 
> now i tell you my problem:
> on a controler raid  (i2o devices) i have a debian distro.
> Recently i have enlarged the logical volumes  on this controler.
> All seemed ok,  i was able to remount the file systems and to copy and install 
> programs and data on resized volumes.
> 
> So, i was out of office and the development server was tuned off by person who 
> wasn't me (so i don't know what is the matter, if happened any thing ).

Are all Physical Volumes accessible (run pvscan and analyze) ?
If any are missing that is likely to explain your problem.

It is hard to help you here without knowing what happened to the system :(

Of course you can use a metadata backup and restore it but this is no
good advice either without knowing which changes have been made (i.e. multiple
lvcreate/lvremove/lvextend/lvreduce meanwhile) and which metadata backup
should be restored, preusmably _all_ PVs are accessible.

You want to read pvcreate and vgcfgrestore manual pages and use
the -ll option of vgcfgrestore to analyze your backup contents.

> 
> later, when in was in office i have turn on the server and the vgscan failed, 
> it said " ..omissis... only  found 499 of 500  LE for LV 
> /dev/volume1/lgv1(0)" .
> I don't undestand what was happened.
> It is possible that a bad tuned off make a problem as this?
> Any body know the answer?
> 
> To solve the trouble i thinked to do in this way:
> to reduce the LV lgv1 of some LE , in this way the info about the number of LE  
> will rewrite and (i hope) the problem vanish.The idea seems good, isn't?
> But i can't do this because the volume group is inactive.
> Anybody know a method to active a volume group also if vgscan fail?
> 
> Please, i MUST to recove the data on the lv, we don't have a backup policy 
> >:^> 
> 
> 
> 
> _______________________________________________
> 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/

-- 

Regards,
Heinz    -- The LVM Guy --

*** 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] i2o devices and lvm Error on LE versus PE

[Gingia]

Alle 11:54, lunedì 22 settembre 2003, Heinz J . Mauelshagen ha scritto:
> Of course you can use a metadata backup and restore it but this is no
> good advice either without knowing which changes have been made (i.e.
> multiple lvcreate/lvremove/lvextend/lvreduce meanwhile) and which metadata
> backup should be restored, preusmably _all_ PVs are accessible.
>
> You want to read pvcreate and vgcfgrestore manual pages and use
> the -ll option of vgcfgrestore to analyze your backup contents.
ThnkYou Very VERY much
I tried to restore the volume with vgcfgrestore but in the past i did it 
wrongly because i read the man pages in bad way (i had a great tears and 
so.... ;^P )
So i didn't see that i used vgcfgrestore  with the option -l so it made the 
list  and not the restore Ah AH AH, if i think that for 2-3 day i "crying" 
the server .....
When you said me to use -l to make a list then i realize.....
Thank you.
In the next time i tell you -if want- what suceeded and i was in that 
situation ....
By