[linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[Joe Harvell]

When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?

I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[Jay Weber]

This problem you're experiencing sounds similar to what I hit after I have
a machine hang while playing about with snapshots.  Did you by chance do
anything with snapshots on w0 volume?

I have machines from beta1, beta2 and beta3 with volumes that are stuck in
the same state that you've run into that I haven't solved yet, granted I
created the problem by messing with snapshots while heavy IO was involved.

On Sun, 28 Jan 2001, Joe Harvell wrote:

> Date: Sun, 28 Jan 2001 12:56:39 -0600
> From: Joe Harvell <harvell@nortelnetworks.com>
> Reply-To: linux-lvm@sistina.com
> To: linux-lvm@sistina.com
> Subject: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get
>     my data)
>
> When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?
>
> I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm
>

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[Joe Harvell]

Jay:

Thanks for responding.  Yes, I have four snapshot LVs in this VG.  However, I don't think there was any I/O at all when I created the snapshot LVs.

So what's up with the af_packet module symbols in the traceback?

Jay Weber wrote:

> This problem you're experiencing sounds similar to what I hit after I have
> a machine hang while playing about with snapshots.  Did you by chance do
> anything with snapshots on w0 volume?
>
> I have machines from beta1, beta2 and beta3 with volumes that are stuck in
> the same state that you've run into that I haven't solved yet, granted I
> created the problem by messing with snapshots while heavy IO was involved.
>
> On Sun, 28 Jan 2001, Joe Harvell wrote:
>
> > Date: Sun, 28 Jan 2001 12:56:39 -0600
> > From: Joe Harvell <harvell@nortelnetworks.com>
> > Reply-To: linux-lvm@sistina.com
> > To: linux-lvm@sistina.com
> > Subject: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get
> >     my data)
> >
> > When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?
> >
> > I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?
> > _______________________________________________
> > linux-lvm mailing list
> > linux-lvm@sistina.com
> > http://lists.sistina.com/mailman/listinfo/linux-lvm
> >
>
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com
> http://lists.sistina.com/mailman/listinfo/linux-lvm

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[lewis]

[-- Attachment #1: Type: text/plain, Size: 1407 bytes --]

On Sun, Jan 28, 2001 at 12:56:39PM -0600, Joe Harvell wrote:
> When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?
> 
> I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?

I believe you want to do a 'vgcfgrestore -nw0 /dev/hda8; vgcfgrestore -nw0
/dev/hdb3'.  Let me know if that fixes things.  And remember, do *not* do a
vgscan right after your vgcfgrestore because it will revert to the previous
state.  Simply run the vgcfgrestore command, and then try the 'vgchange -ay'
to activate the volume group.

Let me know if that fixes things.

Regards,
-- 
AJ Lewis
Sistina Software Inc.                  Voice:  612-379-3951
1313 5th St SE, Suite 111              Fax:    612-379-3952
Minneapolis, MN 55414                  E-Mail: lewis@sistina.com
http://www.sistina.com

Current GPG fingerprint = 3B5F 6011 5216 76A5 2F6B  52A0 941E 1261 0029 2648
Get my key at: http://www.sistina.com/~lewis/gpgkey
 (Unfortunately, the PKS-type keyservers do not work with multiple sub-keys)

-----Begin Obligatory Humorous Quote----------------------------------------
Programming graphics in X is like finding sqrt(pi) using Roman numerals.
-----End Obligatory Humorous Quote------------------------------------------

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[Joe Harvell]

Looks like it didn't work.


[root@btexi0c1 perfect]# vgcfgrestore -n w0 /dev/hda8
vgcfgrestore -- VGDA for "w0" successfully restored to physical volume "/dev/hda8"

[root@btexi0c1 perfect]# vgcfgrestore -n w0 /dev/hdb3
vgcfgrestore -- VGDA for "w0" successfully restored to physical volume "/dev/hdb3"

[root@btexi0c1 perfect]# vgchange -a y
Segmentation fault
[root@btexi0c1 perfect]# tail -30 /var/log/messages
JJan 28 20:15:20 btexi0c1 kernel: LVM version 0.9.1_beta3  by Heinz Mauelshagen  (25/01/2001)
Jan 28 20:15:20 btexi0c1 kernel: lvm -- Module successfully initialized
Jan 28 20:15:35 btexi0c1 kernel: Unable to handle kernel NULL pointer dereference at virtual address 000001b0
Jan 28 20:15:35 btexi0c1 kernel:  printing eip:
Jan 28 20:15:35 btexi0c1 kernel: d88f8e71
Jan 28 20:15:35 btexi0c1 kernel: *pde = 00000000
Jan 28 20:15:35 btexi0c1 kernel: Oops: 0000
Jan 28 20:15:35 btexi0c1 kernel: CPU:    0
Jan 28 20:15:35 btexi0c1 kernel: EIP:    0010:[<d88f8e71>]
Jan 28 20:15:35 btexi0c1 kernel: EFLAGS: 00210246
Jan 28 20:15:35 btexi0c1 kernel: eax: 00000000   ebx: 00000021   ecx: 00000100   edx: 00000100
Jan 28 20:15:35 btexi0c1 kernel: esi: d0dc2000   edi: 00000300   ebp: 00000000   esp: ccaf9b5c
Jan 28 20:15:35 btexi0c1 kernel: ds: 0018   es: 0018   ss: 0018
Jan 28 20:15:35 btexi0c1 kernel: Process vgchange (pid: 31064, stackpage=ccaf9000)
Jan 28 20:15:35 btexi0c1 kernel: Stack: cd609000 d88f90cb d0dc2000 0000ffff 00000000 00000400 ce476200 00000330
Jan 28 20:15:35 btexi0c1 kernel:        00000000 00021c36 d88f77c4 d0dc2000 ce476200 d0dc2000 40239008 00000001
Jan 28 20:15:35 btexi0c1 kernel:        00000100 00000005 ccaf9bdc 00000000 ccaf9d84 d0dc2000 d88f6bda 00000000
Jan 28 20:15:35 btexi0c1 kernel: Call Trace: [<d88f90cb>] [<d88f77c4>] [<d88f6bda>] [af_packet:__insmod_af_packet_O/lib/modules/2.4.0/kernel/net/packet/af+0/96] [<d88f4707>] [do_anonymous_page+50/128] [iget4+194/208]
Jan 28 20:15:35 btexi0c1 kernel:        [update_process_times+32/128] [af_packet:__insmod_af_packet_O/lib/modules/2.4.0/kernel/net/packet/af+-137280/96] [af_packet:__insmod_af_packet_O/lib/modules/2.4.0/kernel/net/packet/af+-137132/96] [timer_bh+531/592] [cached_lookup+16/80] [path_walk+1906/2144] [ret_from_intr+0/32] [chrdev_open+54/64]
Jan 28 20:15:35 btexi0c1 kernel:        [dentry_open+192/336] [filp_open+77/96] [getname+91/160] [sys_ioctl+375/400] [system_call+51/56]
Jan 28 20:15:35 btexi0c1 kernel:
Jan 28 20:15:35 btexi0c1 kernel: Code: 8b 80 b0 01 00 00 5e c3 8d b4 26 00 00 00 00 55 31 d2 57 56


lewis@sistina.com wrote:

> On Sun, Jan 28, 2001 at 12:56:39PM -0600, Joe Harvell wrote:
> > When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?
> >
> > I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?
>
> I believe you want to do a 'vgcfgrestore -nw0 /dev/hda8; vgcfgrestore -nw0
> /dev/hdb3'.  Let me know if that fixes things.  And remember, do *not* do a
> vgscan right after your vgcfgrestore because it will revert to the previous
> state.  Simply run the vgcfgrestore command, and then try the 'vgchange -ay'
> to activate the volume group.
>
> Let me know if that fixes things.
>
> Regards,
> --
> AJ Lewis
> Sistina Software Inc.                  Voice:  612-379-3951
> 1313 5th St SE, Suite 111              Fax:    612-379-3952
> Minneapolis, MN 55414                  E-Mail: lewis@sistina.com
> http://www.sistina.com
>
> Current GPG fingerprint = 3B5F 6011 5216 76A5 2F6B  52A0 941E 1261 0029 2648
> Get my key at: http://www.sistina.com/~lewis/gpgkey
>  (Unfortunately, the PKS-type keyservers do not work with multiple sub-keys)
>
> -----Begin Obligatory Humorous Quote----------------------------------------
> Programming graphics in X is like finding sqrt(pi) using Roman numerals.
> -----End Obligatory Humorous Quote------------------------------------------
>
>                                                   -----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
>    Part 1.2Type: application/pgp-signature

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[lewis]

[-- Attachment #1: Type: text/plain, Size: 1197 bytes --]

On Sun, Jan 28, 2001 at 08:18:24PM -0600, Joe Harvell wrote:
> Looks like it didn't work.

Shoot!  I'll look a this in more detail in the morning when I have access to
my test setup.

-- 
AJ Lewis
Sistina Software Inc.                  Voice:  612-379-3951
1313 5th St SE, Suite 111              Fax:    612-379-3952
Minneapolis, MN 55414                  E-Mail: lewis@sistina.com
http://www.sistina.com

Current GPG fingerprint = 3B5F 6011 5216 76A5 2F6B  52A0 941E 1261 0029 2648
Get my key at: http://www.sistina.com/~lewis/gpgkey
 (Unfortunately, the PKS-type keyservers do not work with multiple sub-keys)

-----Begin Obligatory Humorous Quote----------------------------------------
THE LESSER-KNOWN PROGRAMMING LANGUAGES #16:
C- This language was named for the grade received by its creator when he
submitted it as a class project in a graduate programming class.  C- is
best described as a "low-level" programming language.  In fact, the
language generally requires more C- statements than machine-code statements
to execute a given task.  In this respect, it is very similar to COBOL.
-----End Obligatory Humorous Quote------------------------------------------

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]

Re: [linux-lvm] Follow-up 2: Help: lvm 0.9.1_beta3 oops (can't get my data)

[lewis]

[-- Attachment #1: Type: text/plain, Size: 2033 bytes --]

On Sun, Jan 28, 2001 at 08:06:49PM -0600, lewis@sistina.com wrote:
> On Sun, Jan 28, 2001 at 12:56:39PM -0600, Joe Harvell wrote:
> > When I do a 'vgdisplay -vD w0', I don't run into any problems.  Does this rule out the possibility of a vgcfgrestore helping?
> > 
> > I looked at the man pages for vcfgrestore, but I don't know which destination I should specify.  Is it /dev/w0, or is it 'dev/hda8' and '/dev/hdb3'?
> 
> I believe you want to do a 'vgcfgrestore -nw0 /dev/hda8; vgcfgrestore -nw0
> /dev/hdb3'.  Let me know if that fixes things.  And remember, do *not* do a
> vgscan right after your vgcfgrestore because it will revert to the previous
> state.  Simply run the vgcfgrestore command, and then try the 'vgchange -ay'
> to activate the volume group.

I was told that this is incorrect information by Heinz and that you _should_
run vgscan after running vgcfgrestore because it brings the /etc/lvmtab and
/etc/lvmtab.d/ in sync with the newly restored VGDA.  Sorry about that.  I
was mixed up.

-- 
AJ Lewis
Sistina Software Inc.                  Voice:  612-379-3951
1313 5th St SE, Suite 111              Fax:    612-379-3952
Minneapolis, MN 55414                  E-Mail: lewis@sistina.com
http://www.sistina.com

Current GPG fingerprint = 3B5F 6011 5216 76A5 2F6B  52A0 941E 1261 0029 2648
Get my key at: http://www.sistina.com/~lewis/gpgkey
 (Unfortunately, the PKS-type keyservers do not work with multiple sub-keys)

-----Begin Obligatory Humorous Quote----------------------------------------
THE LESSER-KNOWN PROGRAMMING LANGUAGES #16:
C- This language was named for the grade received by its creator when he
submitted it as a class project in a graduate programming class.  C- is
best described as a "low-level" programming language.  In fact, the
language generally requires more C- statements than machine-code statements
to execute a given task.  In this respect, it is very similar to COBOL.
-----End Obligatory Humorous Quote------------------------------------------

[-- Attachment #2: Type: application/pgp-signature, Size: 232 bytes --]