[linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

hi,

running lvcreate consistently segfaults on me, eg:

 /lib/lvm-iop10/lvcreate -L 600M -n var_cache_lv Main_VG


#0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
    at vg_setup_pointers_for_snapshots.c:37
37	      if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {
(gdb) bt
#0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
    at vg_setup_pointers_for_snapshots.c:37
#1  0x40045d0d in vg_status_with_pv_and_lv (vg_name=0xbffffbe3 "Main_VG", 
    vg=0xbffff254) at vg_status_with_pv_and_lv.c:63
#2  0x0804abbd in strcpy ()
#3  0x400767c7 in __libc_start_main (main=0x8049370 <strcpy+272>, argc=6, 
    ubp_av=0xbffffaa4, init=0x8048ea8 <_init>, fini=0x804beb0 <_fini>, 
    rtld_fini=0x4000db74 <_dl_fini>, stack_end=0xbffffa9c)
    at ../sysdeps/generic/libc-start.c:129
(gdb) print *vg          
$3 = {vg_name = "Main_VG", '\000' <repeats 120 times>, vg_number = 0, 
  vg_access = 3, vg_status = 5, lv_max = 256, lv_cur = 6, lv_open = 5, 
  pv_max = 256, pv_cur = 2, pv_act = 2, dummy = 0, vgda = 0, pe_size = 8192, 
  pe_total = 3690, pe_allocated = 849, pvg_total = 0, proc = 0x0, pv = {
    0xc7c83800, 0xc7c83a00, 0x0 <repeats 255 times>}, lv = {0xc7c83c00, 
    0xc7c83e00, 0xc65e3000, 0xc65e3200, 0xc65e3400, 0xc65e3600, 
    0x0 <repeats 251 times>}, vg_uuid = "Main_VG", '\000' <repeats 25 times>, 
  dummy1 = "\000\000\000`®ÐÇ`­ÐÇ`¬ÐÇ", '\000' <repeats 184 times>}
(gdb) print *lv_chain_ptr
Cannot access memory at address 0xc7c83c00

and boom...

any fixes for this? i've tried compiling with both RH's compat-egcs 
and gcc-2.96-97 compilers, same result each time.

ta,

--paulj

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

doh...

this is 0.9.1beta7, on stock kernel 2.4.9 + LVM 0.9.1beta7 + ext3 + NFS fixups

On Wed, 3 Oct 2001, Paul Jakma wrote:

> hi,
> 
> running lvcreate consistently segfaults on me, eg:
> 
>  /lib/lvm-iop10/lvcreate -L 600M -n var_cache_lv Main_VG
> 
> 
> #0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
>     at vg_setup_pointers_for_snapshots.c:37
> 37	      if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {
> (gdb) bt
> #0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
>     at vg_setup_pointers_for_snapshots.c:37
> #1  0x40045d0d in vg_status_with_pv_and_lv (vg_name=0xbffffbe3 "Main_VG", 
>     vg=0xbffff254) at vg_status_with_pv_and_lv.c:63
> #2  0x0804abbd in strcpy ()
> #3  0x400767c7 in __libc_start_main (main=0x8049370 <strcpy+272>, argc=6, 
>     ubp_av=0xbffffaa4, init=0x8048ea8 <_init>, fini=0x804beb0 <_fini>, 
>     rtld_fini=0x4000db74 <_dl_fini>, stack_end=0xbffffa9c)
>     at ../sysdeps/generic/libc-start.c:129
> (gdb) print *vg          
> $3 = {vg_name = "Main_VG", '\000' <repeats 120 times>, vg_number = 0, 
>   vg_access = 3, vg_status = 5, lv_max = 256, lv_cur = 6, lv_open = 5, 
>   pv_max = 256, pv_cur = 2, pv_act = 2, dummy = 0, vgda = 0, pe_size = 8192, 
>   pe_total = 3690, pe_allocated = 849, pvg_total = 0, proc = 0x0, pv = {
>     0xc7c83800, 0xc7c83a00, 0x0 <repeats 255 times>}, lv = {0xc7c83c00, 
>     0xc7c83e00, 0xc65e3000, 0xc65e3200, 0xc65e3400, 0xc65e3600, 
>     0x0 <repeats 251 times>}, vg_uuid = "Main_VG", '\000' <repeats 25 times>, 
>   dummy1 = "\000\000\000`®ÐÇ`­ÐÇ`¬ÐÇ", '\000' <repeats 184 times>}
> (gdb) print *lv_chain_ptr
> Cannot access memory at address 0xc7c83c00
> 
> and boom...
> 
> any fixes for this? i've tried compiling with both RH's compat-egcs 
> and gcc-2.96-97 compilers, same result each time.
> 
> ta,
> 
> --paulj

--paulj

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

and some more info:

lvm 1.0.1rc3 compiled with RH gcc 2.96-97 running on the kernel
described below (ie 0.9.1b7 kernel) also segfaults in the exact same
way.

regards,

--paulj

On Wed, 3 Oct 2001, Paul Jakma wrote:

> doh...
> 
> this is 0.9.1beta7, on stock kernel 2.4.9 + LVM 0.9.1beta7 + ext3 + NFS fixups
> 
> On Wed, 3 Oct 2001, Paul Jakma wrote:
> 
> > hi,
> > 
> > running lvcreate consistently segfaults on me, eg:
> > 
> >  /lib/lvm-iop10/lvcreate -L 600M -n var_cache_lv Main_VG
> > 
> > 
> > #0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
> >     at vg_setup_pointers_for_snapshots.c:37
> > 37	      if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {
> > (gdb) bt
> > #0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
> >     at vg_setup_pointers_for_snapshots.c:37
> > #1  0x40045d0d in vg_status_with_pv_and_lv (vg_name=0xbffffbe3 "Main_VG", 
> >     vg=0xbffff254) at vg_status_with_pv_and_lv.c:63
> > #2  0x0804abbd in strcpy ()
> > #3  0x400767c7 in __libc_start_main (main=0x8049370 <strcpy+272>, argc=6, 
> >     ubp_av=0xbffffaa4, init=0x8048ea8 <_init>, fini=0x804beb0 <_fini>, 
> >     rtld_fini=0x4000db74 <_dl_fini>, stack_end=0xbffffa9c)
> >     at ../sysdeps/generic/libc-start.c:129
> > (gdb) print *vg          
> > $3 = {vg_name = "Main_VG", '\000' <repeats 120 times>, vg_number = 0, 
> >   vg_access = 3, vg_status = 5, lv_max = 256, lv_cur = 6, lv_open = 5, 
> >   pv_max = 256, pv_cur = 2, pv_act = 2, dummy = 0, vgda = 0, pe_size = 8192, 
> >   pe_total = 3690, pe_allocated = 849, pvg_total = 0, proc = 0x0, pv = {
> >     0xc7c83800, 0xc7c83a00, 0x0 <repeats 255 times>}, lv = {0xc7c83c00, 
> >     0xc7c83e00, 0xc65e3000, 0xc65e3200, 0xc65e3400, 0xc65e3600, 
> >     0x0 <repeats 251 times>}, vg_uuid = "Main_VG", '\000' <repeats 25 times>, 
> >   dummy1 = "\000\000\000`®ÐÇ`­ÐÇ`¬ÐÇ", '\000' <repeats 184 times>}
> > (gdb) print *lv_chain_ptr
> > Cannot access memory at address 0xc7c83c00
> > 
> > and boom...
> > 
> > any fixes for this? i've tried compiling with both RH's compat-egcs 
> > and gcc-2.96-97 compilers, same result each time.
> > 
> > ta,
> > 
> > --paulj
> 
> --paulj
> 
> 
> _______________________________________________
> 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
> 

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[svetljo]

some my thoughts:

you can not expect that lvm tools will work in case the kernel code is 
different version
every time you change the tools you should also patch the kernel

in the case with 0.9.1beta7,
i didn't know that it is possible to use any lvm code prior lvm-1.0.1rc1
( the min -max changes in linus's 2.4.9 -- the reason for lvm-1.0.1rc1)
and thats making me to think that you didn't patch the kernel,
the plain kernel comes with lvm-0.9.1beta2
so see again the begining 


Paul Jakma wrote:

>and some more info:
>
>lvm 1.0.1rc3 compiled with RH gcc 2.96-97 running on the kernel
>described below (ie 0.9.1b7 kernel) also segfaults in the exact same
>way.
>
>regards,
>
>--paulj
>
>On Wed, 3 Oct 2001, Paul Jakma wrote:
>
>>doh...
>>
>>this is 0.9.1beta7, on stock kernel 2.4.9 + LVM 0.9.1beta7 + ext3 + NFS fixups
>>
>>On Wed, 3 Oct 2001, Paul Jakma wrote:
>>
>>>hi,
>>>
>>>running lvcreate consistently segfaults on me, eg:
>>>
>>> /lib/lvm-iop10/lvcreate -L 600M -n var_cache_lv Main_VG
>>>
>>>
>>>#0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
>>>    at vg_setup_pointers_for_snapshots.c:37
>>>37	      if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {
>>>(gdb) bt
>>>#0  0x4004500b in vg_setup_pointers_for_snapshots (vg=0x8058038)
>>>    at vg_setup_pointers_for_snapshots.c:37
>>>#1  0x40045d0d in vg_status_with_pv_and_lv (vg_name=0xbffffbe3 "Main_VG", 
>>>    vg=0xbffff254) at vg_status_with_pv_and_lv.c:63
>>>#2  0x0804abbd in strcpy ()
>>>#3  0x400767c7 in __libc_start_main (main=0x8049370 <strcpy+272>, argc=6, 
>>>    ubp_av=0xbffffaa4, init=0x8048ea8 <_init>, fini=0x804beb0 <_fini>, 
>>>    rtld_fini=0x4000db74 <_dl_fini>, stack_end=0xbffffa9c)
>>>    at ../sysdeps/generic/libc-start.c:129
>>>(gdb) print *vg          
>>>$3 = {vg_name = "Main_VG", '\000' <repeats 120 times>, vg_number = 0, 
>>>  vg_access = 3, vg_status = 5, lv_max = 256, lv_cur = 6, lv_open = 5, 
>>>  pv_max = 256, pv_cur = 2, pv_act = 2, dummy = 0, vgda = 0, pe_size = 8192, 
>>>  pe_total = 3690, pe_allocated = 849, pvg_total = 0, proc = 0x0, pv = {
>>>    0xc7c83800, 0xc7c83a00, 0x0 <repeats 255 times>}, lv = {0xc7c83c00, 
>>>    0xc7c83e00, 0xc65e3000, 0xc65e3200, 0xc65e3400, 0xc65e3600, 
>>>    0x0 <repeats 251 times>}, vg_uuid = "Main_VG", '\000' <repeats 25 times>, 
>>>  dummy1 = "\000\000\000`���`���`���", '\000' <repeats 184 times>}
>>>(gdb) print *lv_chain_ptr
>>>Cannot access memory at address 0xc7c83c00
>>>
>>>and boom...
>>>
>>>any fixes for this? i've tried compiling with both RH's compat-egcs 
>>>and gcc-2.96-97 compilers, same result each time.
>>>
>>>ta,
>>>
>>>--paulj
>>>
>>--paulj
>>
>>
>>_______________________________________________
>>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
>>
>
>
>_______________________________________________
>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
>

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Andreas Dilger]

On Oct 03, 2001  21:12 +0200, svetljo wrote:
> you can not expect that lvm tools will work in case the kernel code is 
> different version
> every time you change the tools you should also patch the kernel

Actually no, you _should_ be able to use different kernel and tools
together _provided_ that the IOP version is the same.  Conversely,
no changes should be made to the tools which break the IOP verion
(either by forcing it to change under 2.4 kernels if it is not needed,
or by changing the data structures and NOT updating the IOP version).

> in the case with 0.9.1beta7,
> i didn't know that it is possible to use any lvm code prior lvm-1.0.1rc1
> ( the min -max changes in linus's 2.4.9 -- the reason for lvm-1.0.1rc1)
> and thats making me to think that you didn't patch the kernel,
> the plain kernel comes with lvm-0.9.1beta2
> so see again the begining 

The min/max issues are easily fixed (and have been reverted in 2.4.10).
Also, the tool version and kernel version SHOULD NOT need to match
exactly (that's what IOP is for).

Infomation about whether the coredump is in user space or a kernel oops
would be helpful.  Also, there is a known issue about the "inline"
functions, which may not have been fixed yet (it is a compiler bug).

Cheers, Andreas
--
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[svetljo]

>
>Actually no, you _should_ be able to use different kernel and tools
>together _provided_ that the IOP version is the same.  Conversely,
>no changes should be made to the tools which break the IOP verion
>(either by forcing it to change under 2.4 kernels if it is not needed,
>or by changing the data structures and NOT updating the IOP version).
>
so i'm wrong

>>in the case with 0.9.1beta7,
>>i didn't know that it is possible to use any lvm code prior lvm-1.0.1rc1
>>( the min -max changes in linus's 2.4.9 -- the reason for lvm-1.0.1rc1)
>>and thats making me to think that you didn't patch the kernel,
>>the plain kernel comes with lvm-0.9.1beta2
>>so see again the begining 
>>
>
>The min/max issues are easily fixed (and have been reverted in 2.4.10).
>Also, the tool version and kernel version SHOULD NOT need to match
>exactly (that's what IOP is for).
>
i ment he could not patch cleanly vanila 2.4.9 with lvm-0.9.1beta7, and 
he is using beta7 tools with beta2 kernel code

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

On Wed, 3 Oct 2001, svetljo wrote:

> i ment he could not patch cleanly vanila 2.4.9 with lvm-0.9.1beta7, and 
> he is using beta7 tools with beta2 kernel code

see my 2nd mail. it's 2.4.9 patched with beta7:

[root@dunlop tmp]# cat /proc/lvm/global 
LVM driver version 0.9.1_beta7 (10/04/2001)

the 1.0.1rc3 tools should be compatible, in fact they are, cause 
vgscan and vgdisplay work fine.

--paulj

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[svetljo]

>
>
>>i ment he could not patch cleanly vanila 2.4.9 with lvm-0.9.1beta7, and 
>>he is using beta7 tools with beta2 kernel code
>>
>
>see my 2nd mail. it's 2.4.9 patched with beta7:
>
my mistake

>
>[root@dunlop tmp]# cat /proc/lvm/global 
>LVM driver version 0.9.1_beta7 (10/04/2001)
>
>the 1.0.1rc3 tools should be compatible, in fact they are, cause 
>vgscan and vgdisplay work fine.
>
they should, but  i had a problem some time ago with lvm-1.0 kernel code 
and lvm-1.0.1rc2 tools
xxscan and xxdisplay worked but when i tried to make a VG/LV to change 
the name or the size
it didn't work, the tools segfaultet 

>
>--paulj
>
>
>_______________________________________________
>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
>

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

On Wed, 3 Oct 2001, Andreas Dilger wrote:

> The min/max issues are easily fixed

yep. fixed it by hand.

> Infomation about whether the coredump is in user space or a kernel oops
> would be helpful.

no kernel oops. purely a segfault in lvcreate.

>  Also, there is a known issue about the "inline"
> functions, which may not have been fixed yet (it is a compiler bug).
> 
> Cheers, Andreas
> --

regards

--paulj

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

On Wed, 3 Oct 2001, Andreas Dilger wrote:

> would be helpful.  Also, there is a known issue about the "inline"
> functions, which may not have been fixed yet (it is a compiler bug).

oh yes..

i read through the archives before posting, and found reference to
this.  however, it happens when compiled with egcs too.

also, the other bug was in was in lv_create(), my one is in
a different place (see subject line).

> Cheers, Andreas

regards,
-- 
Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
Fortune:
Liberty is always dangerous, but it is the safest thing we have.
		-- Harry Emerson Fosdick

[linux-lvm] What have I done!?

[Jason A. Lixfeld]

I was running some kernel version (but I forget which one because I
deleted it! :/  Now, I'm running 2.4.10.  I was running LVM 0.9B7 or B8.
Don't remember.  I upgraded to 1.0 without checking for backwards
compatibility and now my life is a living hell!  I then upgraded to
1.0.1-rc4 in hopes that it would help.  Here's a rundown of my current
confusion as a result of my stupidity:

# /sbin/lvscan
lvscan -- ERROR: VGDA in kernel and lvmtab are NOT consistent; please
run vgscan

# /sbin/vgscan
vgscan -- reading all physical volumes (this may take a while...)
vgscan -- found active volume group "HiTiSO3"
vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully created
vgscan -- WARNING: This program does not do a VGDA backup of your volume
group

(BTW:  whenever I run vgscan, I get these errors all over the place.
The drives are for an LV that is in /proc but not in any of the configs!
I even deleted the /dev file hoping it would kill it)
hdj: read_intr: error=0x04 { DriveStatusError }
hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error
}
hdj: read_intr: error=0x04 { DriveStatusError }
hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error
}
hdl: read_intr: error=0x04 { DriveStatusError }
hdl: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error
}
hdl: read_intr: error=0x04 { DriveStatusError }
hdl: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error
}

# /sbin/pvscan
pvscan -- reading all physical volumes (this may take a while...)
pvscan -- inactive PV "/dev/hdm"  is in no VG  [76.34 GB]
pvscan -- ACTIVE   PV "/dev/hdn1" of VG "foo3" [76.32 GB / 0 free]
pvscan -- inactive PV "/dev/hdi"  is in no VG  [76.34 GB]
pvscan -- ACTIVE   PV "/dev/hdg1" of VG "foo3" [76.33 GB / 0 free]
pvscan -- inactive PV "/dev/hde"  is in no VG  [4.50 MB]
pvscan -- total: 5 [305.34 GB] / in use: 2 [152.66 GB] / in no VG: 3
[152.67 GB]

# cat /proc/lvm/global
LVM driver LVM version 1.0.1-rc4(03/10/2001)

Total:  2 VGs  4 PVs  2 LVs (1 LV open 1 times)

Global: 1244191 bytes malloced   IOP version: 10   5:22:56 active

VG:  foo3  [2 PV, 1 LV/1 open]  PE Size: 4096 KB
  Usage [KB/PE]: 160071680 /39080 total  160071680 /39080 used  0 /0
free
  PVs: [AA] hdn1                  80031744 /19539   80031744 /19539
0 /0     
       [AA] hdg1                  80039936 /19541   80039936 /19541
0 /0     
    LV:  [AWDL  ] FOO3                     160071680 /39080    1x open

VG:  foo1  [2 PV, 1 LV/0 open]  PE Size: 4096 KB
  Usage [KB/PE]: 156295168 /38158 total  156295168 /38158 used  0 /0
free
  PVs: [AA] hdj                   78147584 /19079   78147584 /19079
0 /0     
       [AA] hdl                   78147584 /19079   78147584 /19079
0 /0     
    LV:  [AWDL  ] FOO1                  156295168 /38158    close

# ls /etc/lvmconf
foo3.conf

# more /etc/lvmtab
foo3

# ls /etc/lvmtab.d/
foo3

# /sbin/pvcreate /dev/hdk
pvcreate -- ERROR: VGDA in kernel and lvmtab are NOT consistent; please
run vgscan

# /sbin/pvdisplay        
pvdisplay -- ERROR: VGDA in kernel and lvmtab are NOT consistent; please
run vgscan

# /sbin/vgdisplay
vgdisplay -- ERROR: VGDA in kernel and lvmtab are NOT consistent; please
run vgscan

# /sbin/lvdisplay
lvdisplay -- ERROR: VGDA in kernel and lvmtab are NOT consistent; please
run vgscan

There's a couple things:

1) I deleted everything in /etc/lv* except for anything related to foo3,
yet there is another VG, foo1 in /proc.
2) I can't remove inactive drives from showing up in pvscan

Basically, I just want to delete EVERYTHING EXCEPT for foo3.  I'm going
to try a downgraded kernel with the old programs and the old drivers and
see if that makes any difference (I do have all the configs backed up
for the VGs that I deleted previously).

If anyone has some insight on how I can just delete these things and be
done with it, I'd appreciate it.  I can't run any vg commands without
getting those VGDA consistency errors.  I tried a fix for that which I
found on the archives, but that didn't help at all.

Any info would be greatly appreciated!!

-JL

RE: [linux-lvm] What have I done!?

[Jason A. Lixfeld]

Well, because I don't remember what kernel version I was running or what
version of the LVM code I was running, I think me trying to compile
another kernel with the drivers is going to be very much in vein, so I'm
not going to even bother.

Just like the last message:  If anyone can tell me how go about blowing
away the various LVMs so I can try to re-do this thing, I'd appreciate
it!

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> Sent: October 3, 2001 10:54 PM
> To: linux-lvm@sistina.com
> Subject: [linux-lvm] What have I done!?
> 
> 
> 
> 
> I was running some kernel version (but I forget which one 
> because I deleted it! :/  Now, I'm running 2.4.10.  I was 
> running LVM 0.9B7 or B8. Don't remember.  I upgraded to 1.0 
> without checking for backwards compatibility and now my life 
> is a living hell!  I then upgraded to 1.0.1-rc4 in hopes that 
> it would help.  Here's a rundown of my current confusion as a 
> result of my stupidity:
> 
> # /sbin/lvscan
> lvscan -- ERROR: VGDA in kernel and lvmtab are NOT 
> consistent; please run vgscan
> 
> # /sbin/vgscan
> vgscan -- reading all physical volumes (this may take a 
> while...) vgscan -- found active volume group "HiTiSO3" 
> vgscan -- "/etc/lvmtab" and "/etc/lvmtab.d" successfully 
> created vgscan -- WARNING: This program does not do a VGDA 
> backup of your volume group
> 
> (BTW:  whenever I run vgscan, I get these errors all over the 
> place. The drives are for an LV that is in /proc but not in 
> any of the configs! I even deleted the /dev file hoping it 
> would kill it)
> hdj: read_intr: error=0x04 { DriveStatusError }
> hdj: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> hdj: read_intr: error=0x04 { DriveStatusError }
> hdj: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> hdl: read_intr: error=0x04 { DriveStatusError }
> hdl: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> hdl: read_intr: error=0x04 { DriveStatusError }
> hdl: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> 
> # /sbin/pvscan
> pvscan -- reading all physical volumes (this may take a 
> while...) pvscan -- inactive PV "/dev/hdm"  is in no VG  [76.34 GB]
> pvscan -- ACTIVE   PV "/dev/hdn1" of VG "foo3" [76.32 GB / 0 free]
> pvscan -- inactive PV "/dev/hdi"  is in no VG  [76.34 GB]
> pvscan -- ACTIVE   PV "/dev/hdg1" of VG "foo3" [76.33 GB / 0 free]
> pvscan -- inactive PV "/dev/hde"  is in no VG  [4.50 MB]
> pvscan -- total: 5 [305.34 GB] / in use: 2 [152.66 GB] / in 
> no VG: 3 [152.67 GB]
> 
> # cat /proc/lvm/global
> LVM driver LVM version 1.0.1-rc4(03/10/2001)
> 
> Total:  2 VGs  4 PVs  2 LVs (1 LV open 1 times)
> 
> Global: 1244191 bytes malloced   IOP version: 10   5:22:56 active
> 
> VG:  foo3  [2 PV, 1 LV/1 open]  PE Size: 4096 KB
>   Usage [KB/PE]: 160071680 /39080 total  160071680 /39080 
> used  0 /0 free
>   PVs: [AA] hdn1                  80031744 /19539   80031744 /19539
> 0 /0     
>        [AA] hdg1                  80039936 /19541   80039936 /19541
> 0 /0     
>     LV:  [AWDL  ] FOO3                     160071680 /39080    1x open
> 
> VG:  foo1  [2 PV, 1 LV/0 open]  PE Size: 4096 KB
>   Usage [KB/PE]: 156295168 /38158 total  156295168 /38158 
> used  0 /0 free
>   PVs: [AA] hdj                   78147584 /19079   78147584 /19079
> 0 /0     
>        [AA] hdl                   78147584 /19079   78147584 /19079
> 0 /0     
>     LV:  [AWDL  ] FOO1                  156295168 /38158    close
> 
> # ls /etc/lvmconf
> foo3.conf
> 
> # more /etc/lvmtab
> foo3
> 
> # ls /etc/lvmtab.d/
> foo3
> 
> # /sbin/pvcreate /dev/hdk
> pvcreate -- ERROR: VGDA in kernel and lvmtab are NOT 
> consistent; please run vgscan
> 
> # /sbin/pvdisplay        
> pvdisplay -- ERROR: VGDA in kernel and lvmtab are NOT 
> consistent; please run vgscan
> 
> # /sbin/vgdisplay
> vgdisplay -- ERROR: VGDA in kernel and lvmtab are NOT 
> consistent; please run vgscan
> 
> # /sbin/lvdisplay
> lvdisplay -- ERROR: VGDA in kernel and lvmtab are NOT 
> consistent; please run vgscan
> 
> There's a couple things:
> 
> 1) I deleted everything in /etc/lv* except for anything 
> related to foo3, yet there is another VG, foo1 in /proc.
> 2) I can't remove inactive drives from showing up in pvscan
> 
> Basically, I just want to delete EVERYTHING EXCEPT for foo3.  
> I'm going to try a downgraded kernel with the old programs 
> and the old drivers and see if that makes any difference (I 
> do have all the configs backed up for the VGs that I deleted 
> previously).
> 
> If anyone has some insight on how I can just delete these 
> things and be done with it, I'd appreciate it.  I can't run 
> any vg commands without getting those VGDA consistency 
> errors.  I tried a fix for that which I found on the 
> archives, but that didn't help at all.
> 
> Any info would be greatly appreciated!!
> 
> -JL
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

Re: [linux-lvm] What have I done!?

[idsfa]

On Wed, Oct 03, 2001 at 10:54:21PM -0400, Jason A. Lixfeld wrote:

> hdj: read_intr: error=0x04 { DriveStatusError }
> hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
> hdj: read_intr: error=0x04 { DriveStatusError }
> hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
> hdl: read_intr: error=0x04 { DriveStatusError }
> hdl: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
> hdl: read_intr: error=0x04 { DriveStatusError }
> hdl: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }

Those are IDE driver errors.  Judging from the drive devices, you
have quite a few ide controllers.  Did you compile in the drivers
for all of them?  Some of the cards (and even some on-board chipsets)
are NOT supported by default.  In particular, Promise & HiPoint
cards require special kernel compile time options.

What do you see in /proc/ide?

Note that if you compiled the drivers as modules, those modules
MUST be loaded BEFORE lvm is started.

  Michael Kellen
  $ fortune -m Kellen

RE: [linux-lvm] What have I done!?

[Jason A. Lixfeld]

Hi Michael,

	All of the cards are Promise cards.  One is a FastTrack100 TX2
and two others are Promise Ultra100s:

  Bus  0, device   8, function  0:
    RAID bus controller: PCI device 105a:6268 (Promise Technology, Inc.)
(rev 1).
      IRQ 15.
      Master Capable.  Latency=64.  Min Gnt=4.Max Lat=18.
      I/O at 0xa400 [0xa407].
      I/O at 0xa800 [0xa803].
      I/O at 0xac00 [0xac07].
      I/O at 0xb000 [0xb003].
      I/O at 0xb400 [0xb40f].
      Non-prefetchable 32 bit memory at 0xec040000 [0xec04ffff].
  Bus  0, device   9, function  0:
    Unknown mass storage controller: Promise Technology, Inc. 20267 (rev
2).
      IRQ 10.
      Master Capable.  Latency=64.  
      I/O at 0xc000 [0xc007].
      I/O at 0xc400 [0xc403].
      I/O at 0xb800 [0xb807].
      I/O at 0xbc00 [0xbc03].
      I/O at 0xc800 [0xc83f].
      Non-prefetchable 32 bit memory at 0xec020000 [0xec03ffff].
  Bus  0, device  10, function  0:
    Unknown mass storage controller: Promise Technology, Inc. 20267 (#2)
(rev 2).
      IRQ 12.
      Master Capable.  Latency=64.  
      I/O at 0xd400 [0xd407].
      I/O at 0xd800 [0xd803].
      I/O at 0xcc00 [0xcc07].
      I/O at 0xd000 [0xd003].
      I/O at 0xdc00 [0xdc3f].
      Non-prefetchable 32 bit memory at 0xec000000 [0xec01ffff].

# cat /usr/src/linux/.config
...
CONFIG_BLK_DEV_IDE=y
CONFIG_BLK_DEV_IDEDISK=y
CONFIG_IDEDISK_MULTI_MODE=y
CONFIG_BLK_DEV_IDECD=y
CONFIG_BLK_DEV_CMD640=y
CONFIG_BLK_DEV_RZ1000=y
CONFIG_BLK_DEV_IDEPCI=y
CONFIG_IDEPCI_SHARE_IRQ=y
CONFIG_BLK_DEV_IDEDMA_PCI=y
CONFIG_BLK_DEV_ADMA=y
CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_BLK_DEV_IDEDMA=y
CONFIG_BLK_DEV_PIIX=y
CONFIG_PIIX_TUNING=y
CONFIG_BLK_DEV_PDC202XX=y
CONFIG_PDC202XX_BURST=y
CONFIG_PDC202XX_FORCE=y
CONFIG_IDEDMA_AUTO=y
CONFIG_BLK_DEV_IDE_MODES=y
CONFIG_BLK_DEV_ATARAID=y
CONFIG_BLK_DEV_ATARAID_PDC=y
...

Everything is compiled right into the kernel, no modules.

$ ls -al /proc/ide/
total 0
dr-xr-xr-x    8 root     root            0 Oct  4 06:58 .
dr-xr-xr-x   47 root     root            0 Oct  3 17:12 ..
-r--r--r--    1 root     root            0 Oct  4 06:58 drivers
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hda -> ide0/hda
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hde -> ide2/hde
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdf -> ide2/hdf
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdg -> ide3/hdg
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdh -> ide3/hdh
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdi -> ide4/hdi
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdj -> ide4/hdj
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdk -> ide5/hdk
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdl -> ide5/hdl
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdm -> ide6/hdm
lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdn -> ide6/hdn
dr-xr-xr-x    3 root     root            0 Oct  4 06:58 ide0
dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide2
dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide3
dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide4
dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide5
dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide6
-r--r--r--    1 root     root            0 Oct  4 06:58 pdc202xx
-r--r--r--    1 root     root            0 Oct  4 06:58 piix
$

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of idsfa@visi.com
> Sent: October 4, 2001 1:44 AM
> To: linux-lvm@sistina.com
> Subject: Re: [linux-lvm] What have I done!?
> 
> 
> 
> On Wed, Oct 03, 2001 at 10:54:21PM -0400, Jason A. Lixfeld wrote:
> 
> > hdj: read_intr: error=0x04 { DriveStatusError }
> > hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest 
> > Error }
> > hdj: read_intr: error=0x04 { DriveStatusError }
> > hdj: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> > hdl: read_intr: error=0x04 { DriveStatusError }
> > hdl: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> > hdl: read_intr: error=0x04 { DriveStatusError }
> > hdl: read_intr: status=0x59 { DriveReady SeekComplete 
> DataRequest Error }
> 
> Those are IDE driver errors.  Judging from the drive devices, 
> you have quite a few ide controllers.  Did you compile in the 
> drivers for all of them?  Some of the cards (and even some 
> on-board chipsets) are NOT supported by default.  In 
> particular, Promise & HiPoint cards require special kernel 
> compile time options.
> 
> What do you see in /proc/ide?
> 
> Note that if you compiled the drivers as modules, those 
> modules MUST be loaded BEFORE lvm is started.
> 
>   Michael Kellen
>   $ fortune -m Kellen
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

RE: [linux-lvm] What have I done!?

[Jason A. Lixfeld]

Is there a way that I can blow the whole thing away and start over?  I
think I've screwed things up too badly to try to resurrect them.  I'd
really like to blow it all away and start over and see if this stuff
happens again, but again I can't use the lv* tools to remove the
configurations because of the errors I mention in my initial post so I
need a manual way to blow away the entire LVM configuration and start
from scratch again.

Plz help! :P

BTW:  The Seg Fault issue is on another machine..

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> Sent: October 4, 2001 7:00 AM
> To: linux-lvm@sistina.com
> Subject: RE: [linux-lvm] What have I done!?
> 
> 
> 
> 
> Hi Michael,
> 
> 	All of the cards are Promise cards.  One is a 
> FastTrack100 TX2 and two others are Promise Ultra100s:
> 
>   Bus  0, device   8, function  0:
>     RAID bus controller: PCI device 105a:6268 (Promise 
> Technology, Inc.) (rev 1).
>       IRQ 15.
>       Master Capable.  Latency=64.  Min Gnt=4.Max Lat=18.
>       I/O at 0xa400 [0xa407].
>       I/O at 0xa800 [0xa803].
>       I/O at 0xac00 [0xac07].
>       I/O at 0xb000 [0xb003].
>       I/O at 0xb400 [0xb40f].
>       Non-prefetchable 32 bit memory at 0xec040000 [0xec04ffff].
>   Bus  0, device   9, function  0:
>     Unknown mass storage controller: Promise Technology, Inc. 
> 20267 (rev 2).
>       IRQ 10.
>       Master Capable.  Latency=64.  
>       I/O at 0xc000 [0xc007].
>       I/O at 0xc400 [0xc403].
>       I/O at 0xb800 [0xb807].
>       I/O at 0xbc00 [0xbc03].
>       I/O at 0xc800 [0xc83f].
>       Non-prefetchable 32 bit memory at 0xec020000 [0xec03ffff].
>   Bus  0, device  10, function  0:
>     Unknown mass storage controller: Promise Technology, Inc. 
> 20267 (#2) (rev 2).
>       IRQ 12.
>       Master Capable.  Latency=64.  
>       I/O at 0xd400 [0xd407].
>       I/O at 0xd800 [0xd803].
>       I/O at 0xcc00 [0xcc07].
>       I/O at 0xd000 [0xd003].
>       I/O at 0xdc00 [0xdc3f].
>       Non-prefetchable 32 bit memory at 0xec000000 [0xec01ffff].
> 
> # cat /usr/src/linux/.config
> ...
> CONFIG_BLK_DEV_IDE=y
> CONFIG_BLK_DEV_IDEDISK=y
> CONFIG_IDEDISK_MULTI_MODE=y
> CONFIG_BLK_DEV_IDECD=y
> CONFIG_BLK_DEV_CMD640=y
> CONFIG_BLK_DEV_RZ1000=y
> CONFIG_BLK_DEV_IDEPCI=y
> CONFIG_IDEPCI_SHARE_IRQ=y
> CONFIG_BLK_DEV_IDEDMA_PCI=y
> CONFIG_BLK_DEV_ADMA=y
> CONFIG_IDEDMA_PCI_AUTO=y
> CONFIG_BLK_DEV_IDEDMA=y
> CONFIG_BLK_DEV_PIIX=y
> CONFIG_PIIX_TUNING=y
> CONFIG_BLK_DEV_PDC202XX=y
> CONFIG_PDC202XX_BURST=y
> CONFIG_PDC202XX_FORCE=y
> CONFIG_IDEDMA_AUTO=y
> CONFIG_BLK_DEV_IDE_MODES=y
> CONFIG_BLK_DEV_ATARAID=y
> CONFIG_BLK_DEV_ATARAID_PDC=y
> ...
> 
> Everything is compiled right into the kernel, no modules.
> 
> $ ls -al /proc/ide/
> total 0
> dr-xr-xr-x    8 root     root            0 Oct  4 06:58 .
> dr-xr-xr-x   47 root     root            0 Oct  3 17:12 ..
> -r--r--r--    1 root     root            0 Oct  4 06:58 drivers
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hda 
> -> ide0/hda
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hde 
> -> ide2/hde
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdf 
> -> ide2/hdf
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdg 
> -> ide3/hdg
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdh 
> -> ide3/hdh
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdi 
> -> ide4/hdi
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdj 
> -> ide4/hdj
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdk 
> -> ide5/hdk
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdl 
> -> ide5/hdl
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdm 
> -> ide6/hdm
> lrwxrwxrwx    1 root     root            8 Oct  4 06:58 hdn 
> -> ide6/hdn
> dr-xr-xr-x    3 root     root            0 Oct  4 06:58 ide0
> dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide2
> dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide3
> dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide4
> dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide5
> dr-xr-xr-x    4 root     root            0 Oct  4 06:58 ide6
> -r--r--r--    1 root     root            0 Oct  4 06:58 pdc202xx
> -r--r--r--    1 root     root            0 Oct  4 06:58 piix
> $
> 
> > -----Original Message-----
> > From: linux-lvm-admin@sistina.com
> > [mailto:linux-lvm-admin@sistina.com] On Behalf Of idsfa@visi.com
> > Sent: October 4, 2001 1:44 AM
> > To: linux-lvm@sistina.com
> > Subject: Re: [linux-lvm] What have I done!?
> > 
> > 
> > 
> > On Wed, Oct 03, 2001 at 10:54:21PM -0400, Jason A. Lixfeld wrote:
> > 
> > > hdj: read_intr: error=0x04 { DriveStatusError }
> > > hdj: read_intr: status=0x59 { DriveReady SeekComplete DataRequest
> > > Error }
> > > hdj: read_intr: error=0x04 { DriveStatusError }
> > > hdj: read_intr: status=0x59 { DriveReady SeekComplete 
> > DataRequest Error }
> > > hdl: read_intr: error=0x04 { DriveStatusError }
> > > hdl: read_intr: status=0x59 { DriveReady SeekComplete
> > DataRequest Error }
> > > hdl: read_intr: error=0x04 { DriveStatusError }
> > > hdl: read_intr: status=0x59 { DriveReady SeekComplete
> > DataRequest Error }
> > 
> > Those are IDE driver errors.  Judging from the drive devices,
> > you have quite a few ide controllers.  Did you compile in the 
> > drivers for all of them?  Some of the cards (and even some 
> > on-board chipsets) are NOT supported by default.  In 
> > particular, Promise & HiPoint cards require special kernel 
> > compile time options.
> > 
> > What do you see in /proc/ide?
> > 
> > Note that if you compiled the drivers as modules, those
> > modules MUST be loaded BEFORE lvm is started.
> > 
> >   Michael Kellen
> >   $ fortune -m Kellen
> > 
> > _______________________________________________
> > linux-lvm mailing list
> > linux-lvm@sistina.com
> > http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> > read the 
> > LVM HOW-TO at 
> > http://www.sistina.com/lvm/Pages/howto.html
> > 
> 
> 
> _______________________________________________
> 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
> 

Re: [linux-lvm] What have I done!?

[Andreas Dilger]

On Oct 04, 2001  07:24 -0400, Jason A. Lixfeld wrote:
> Is there a way that I can blow the whole thing away and start over?  I
> think I've screwed things up too badly to try to resurrect them.  I'd
> really like to blow it all away and start over and see if this stuff
> happens again, but again I can't use the lv* tools to remove the
> configurations because of the errors I mention in my initial post so I
> need a manual way to blow away the entire LVM configuration and start
> from scratch again.

"pvcreate -ff" I think

Cheers, Andreas
--
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert

Re: [linux-lvm] What have I done!?

[Steven Lembark]

-- Andreas Dilger <adilger@turbolabs.com>

>> Is there a way that I can blow the whole thing away and start over?  I
>> think I've screwed things up too badly to try to resurrect them.  I'd
>> really like to blow it all away and start over and see if this stuff
>> happens again, but again I can't use the lv* tools to remove the
>> configurations because of the errors I mention in my initial post so I
>> need a manual way to blow away the entire LVM configuration and start
>> from scratch again.
>
> "pvcreate -ff" I think

If you really want to wipe things clean:

    (1) vgexport /dev/vgname; # avoids collisions w/ stale config info.
    (2) for disk in your disks here; do pvcreate -ff $disk; done;

The export isn't strictly necessar:  rm -rf and a vgscan should do the
same basic thing with a few nastygrams added in.  Personally, I get
enough nastygrams to prefer avoiding a few more :-)


--
Steven Lembark                                               2930 W. Palmer
Workhorse Computing                                       Chicago, IL 60647
                                                            +1 800 762 1582

Re: [linux-lvm] What have I done!?

[Heinz J . Mauelshagen]

On Thu, Oct 04, 2001 at 12:57:21PM -0500, Steven Lembark wrote:
> 
> 
> -- Andreas Dilger <adilger@turbolabs.com>
> 
> >> Is there a way that I can blow the whole thing away and start over?  I
> >> think I've screwed things up too badly to try to resurrect them.  I'd
> >> really like to blow it all away and start over and see if this stuff
> >> happens again, but again I can't use the lv* tools to remove the
> >> configurations because of the errors I mention in my initial post so I
> >> need a manual way to blow away the entire LVM configuration and start
> >> from scratch again.
> >
> > "pvcreate -ff" I think
> 
> If you really want to wipe things clean:
> 
>     (1) vgexport /dev/vgname; # avoids collisions w/ stale config info.
>     (2) for disk in your disks here; do pvcreate -ff $disk; done;
> 
> The export isn't strictly necessar:  rm -rf and a vgscan should do the
> same basic thing with a few nastygrams added in.  Personally, I get
> enough nastygrams to prefer avoiding a few more :-)

I'm curious: which ones are thouse?

> 
> 
> --
> Steven Lembark                                               2930 W. Palmer
> Workhorse Computing                                       Chicago, IL 60647
>                                                             +1 800 762 1582
> 
> _______________________________________________
> 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

-- 

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] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

I'm having the same problem:

# /sbin/lvcreate -L 111G -n test_lv test_vg
Segmentation fault (core dumped)
#

Running:
LVM driver LVM version 1.0.1-rc4(03/10/2001)
Kernel 2.4.10
ext3-2.4-0.9.10-2410

What's the fix here?!

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Paul Jakma
> Sent: October 3, 2001 8:46 PM
> To: linux-lvm@sistina.com
> Subject: Re: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> On Wed, 3 Oct 2001, Andreas Dilger wrote:
> 
> > would be helpful.  Also, there is a known issue about the "inline" 
> > functions, which may not have been fixed yet (it is a compiler bug).
> 
> oh yes..
> 
> i read through the archives before posting, and found 
> reference to this.  however, it happens when compiled with egcs too.
> 
> also, the other bug was in was in lv_create(), my one is in
> a different place (see subject line).
> 
> > Cheers, Andreas
> 
> regards,
> -- 
> Paul Jakma	paul@clubi.ie	paul@jakma.org	Key ID: 64A2FF6A
> Fortune:
> Liberty is always dangerous, but it is the safest thing we have.
> 		-- Harry Emerson Fosdick
> 
> 
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Andreas Dilger]

On Oct 03, 2001  23:28 -0400, Jason A. Lixfeld wrote:
> I'm having the same problem:
> 
> # /sbin/lvcreate -L 111G -n test_lv test_vg
> Segmentation fault (core dumped)
> #
> 
> Running:
> LVM driver LVM version 1.0.1-rc4(03/10/2001)
> Kernel 2.4.10
> ext3-2.4-0.9.10-2410
> 
> What's the fix here?!

Uh, gdb /sbin/lvcreate (or maybe /lib/lvm-iop10/lvcreate), then
gdb> r -v -d -L 111G -n test_lv test_vg
[lots of stuff]
Segmentation fault (core dumped)
gdb> info stack

Cheers, Andreas
--
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert

RE: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

Wow!  Lots and LOTS of stuff!!  Didn't think it was ever going to end!!

Here's what yer looking for:

#0  lv_create (vg=0x8, lv=0xbffffb74, 
    lv_name=0xbffffb98
"¥üÿ¿®üÿ¿¿üÿ¿áüÿ¿îüÿ¿±þÿ¿Ðþÿ¿íþÿ¿\002ÿÿ¿!ÿÿ¿,ÿÿ¿<ÿÿ¿Dÿÿ¿Tÿÿ¿bÿÿ¿sÿÿ¿\201
ÿÿ¿\214ÿÿ¿\227ÿÿ¿Úÿÿ¿")
    at lv_create_remove.c:42
#1  0x0804b03d in strcpy ()@../sysdeps/generic/strcpy.c:31
#2  0x4006c177 in __libc_start_main (main=0x8049390 <strcpy+384>,
argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>, 
    fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Andreas Dilger
> Sent: October 3, 2001 11:34 PM
> To: linux-lvm@sistina.com
> Subject: Re: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> On Oct 03, 2001  23:28 -0400, Jason A. Lixfeld wrote:
> > I'm having the same problem:
> > 
> > # /sbin/lvcreate -L 111G -n test_lv test_vg
> > Segmentation fault (core dumped)
> > #
> > 
> > Running:
> > LVM driver LVM version 1.0.1-rc4(03/10/2001)
> > Kernel 2.4.10
> > ext3-2.4-0.9.10-2410
> > 
> > What's the fix here?!
> 
> Uh, gdb /sbin/lvcreate (or maybe /lib/lvm-iop10/lvcreate), then
> gdb> r -v -d -L 111G -n test_lv test_vg
> [lots of stuff]
> Segmentation fault (core dumped)
> gdb> info stack
> 
> Cheers, Andreas
> --
> Andreas Dilger  \ "If a man ate a pound of pasta and a pound 
> of antipasto,
>                  \  would they cancel out, leaving him still hungry?"
> http://www-mddsp.enel.ucalgary.ca/People/adilger/             
>   -- Dogbert
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Andreas Dilger]

On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> Wow!  Lots and LOTS of stuff!!  Didn't think it was ever going to end!!

Some of it is even useful.  Maybe you can include some of the stuff at
the end, or put it on a web page somewhere.

> Here's what yer looking for:
> 
> #0  lv_create (vg=0x8, lv=0xbffffb74, 
>     lv_name=0xbffffb98
> "????????????????????????????????\002???!???,???<???D???T???b???s???\201
> ???\214???\227???????")
>     at lv_create_remove.c:42
> #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31

So it is dereferencing an uninitialized string (or so it appears).  It may
be that the string is corrupted elsewhere, however.  I'm surprised that
lv_check_name() in lv_create_remove() didn't catch this.  However, this
may also be a red herring, since sometimes GDB shows junk.

It is also suspicious because the lv and lv_name pointers are 0x24 apart,
which shouldn't be the case since lv_name is the first item in lv_t.

> #2  0x4006c177 in __libc_start_main (main=0x8049390 <strcpy+384>,
> argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>, 
>     fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
> stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129

If you try running with "export MALLOC_CHECK_=2 gdb <path>/lvcreate",
it may give more indication as to where the actual corruption is happening.

I just recompiled the tools, and did not have any problems with lvcreate,
so it must be something strange.

Cheers, Andreas
--
Andreas Dilger  \ "If a man ate a pound of pasta and a pound of antipasto,
                 \  would they cancel out, leaving him still hungry?"
http://www-mddsp.enel.ucalgary.ca/People/adilger/               -- Dogbert

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Sean Burford]

This is exactly what I was seeing.  I replaced the lv_create call with
an lv_create_remove call and it worked.

Note that not only lv_name is corrupt, but so are the vg and lv.

I was told that this was due to a buggy compiler, though I have not had
time to find and use a good compiler yet.  I guess it is only a metter
of time until something else crashes, since I comiled my kernel, modules
and LVM with the compiler :)  I'll let you know when I find a comilper
that works.  Andreas, what compiler version are you using?

Andreas Dilger wrote:
> 
> On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> > Here's what yer looking for:
> >
> > #0  lv_create (vg=0x8, lv=0xbffffb74,
> >     lv_name=0xbffffb98
> > "????????????????????????????????\002???!???,???<???D???T???b???s???\201
> > ???\214???\227???????")
> >     at lv_create_remove.c:42
> > #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31
> 
> So it is dereferencing an uninitialized string (or so it appears).  It may
> be that the string is corrupted elsewhere, however.  I'm surprised that
> lv_check_name() in lv_create_remove() didn't catch this.  However, this
> may also be a red herring, since sometimes GDB shows junk.
> 
> It is also suspicious because the lv and lv_name pointers are 0x24 apart,
> which shouldn't be the case since lv_name is the first item in lv_t.

-- 
Sean Burford    x34135
ITS Systems Specialist
Adelaide University

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

ok... looking at tools/lib/vg_setup_pointers_for_snapshots():

 if ( vg == NULL) goto vg_setup_pointers_for_snapshots_end;
   for ( l = 0; l < vg->lv_max; l++) {
      lv_t *lv_chain_ptr = vg->lv[l];
      if ( lv_chain_ptr == NULL) continue;
      if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {

isn't this a wee bit suspect? we test for NULL /after/ the assignment, 
and continue anyway? next line deferences that pointer, irregardless 
of the test for previously..

anyway... this isn't my problem, vg->lv_max[1] / lv_chain_ptr is not 
null, it's invalid.

anyone know how to get a longer backtrace from gdb?

--paulj

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Joe Thornber]

On Thu, Oct 04, 2001 at 11:31:15AM +0100, Paul Jakma wrote:
> ok... looking at tools/lib/vg_setup_pointers_for_snapshots():
> 
>  if ( vg == NULL) goto vg_setup_pointers_for_snapshots_end;
>    for ( l = 0; l < vg->lv_max; l++) {
>       lv_t *lv_chain_ptr = vg->lv[l];
>       if ( lv_chain_ptr == NULL) continue;
>       if ( lv_chain_ptr->lv_access & LV_SNAPSHOT_ORG) {
> 
> isn't this a wee bit suspect? we test for NULL /after/ the assignment, 
> and continue anyway? next line deferences that pointer, irregardless 
> of the test for previously..

No, this snippet looks right to me.  How can we test before the assignment ?
If lv_chain_ptr is NULL the code then jumps back to the top of the 'for' loop.

- Joe

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Paul Jakma]

On Thu, 4 Oct 2001, Joe Thornber wrote:

> >    for ( l = 0; l < vg->lv_max; l++) {
> >       lv_t *lv_chain_ptr = vg->lv[l];
> >       if ( lv_chain_ptr == NULL) continue;

> No, this snippet looks right to me.  How can we test before the assignment ?

well, it's not a biggy, but you could test vg->lv[l] for NULL and 
conditionally assign and save a copy if NULL.

in light of your next sentence though, doesn't matter.

> If lv_chain_ptr is NULL the code then jumps back to the top of the
> 'for' loop.

ah... Doh! of course.

> - Joe

anyway, any idea how to get a deeper back trace? or does info stack/bt 
print out the complete stack trace?

--paulj

RE: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

Ok, done...

http://www.andromedas.com/lvm/session.log

It's 35MB and I'm only on a 3MB DSL connection so it's gonna take a
while to load! :)

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Andreas Dilger
> Sent: October 4, 2001 2:05 AM
> To: linux-lvm@sistina.com
> Subject: Re: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> > Wow!  Lots and LOTS of stuff!!  Didn't think it was ever going to 
> > end!!
> 
> Some of it is even useful.  Maybe you can include some of the 
> stuff at the end, or put it on a web page somewhere.
> 
> > Here's what yer looking for:
> > 
> > #0  lv_create (vg=0x8, lv=0xbffffb74, 
> >     lv_name=0xbffffb98 
> > 
> "????????????????????????????????\002???!???,???<???D???T???b???s???\2
> > 01
> > ???\214???\227???????")
> >     at lv_create_remove.c:42
> > #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31
> 
> So it is dereferencing an uninitialized string (or so it 
> appears).  It may be that the string is corrupted elsewhere, 
> however.  I'm surprised that
> lv_check_name() in lv_create_remove() didn't catch this.  
> However, this may also be a red herring, since sometimes GDB 
> shows junk.
> 
> It is also suspicious because the lv and lv_name pointers are 
> 0x24 apart, which shouldn't be the case since lv_name is the 
> first item in lv_t.
> 
> > #2  0x4006c177 in __libc_start_main (main=0x8049390 <strcpy+384>, 
> > argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>,
> >     fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
> > stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129
> 
> If you try running with "export MALLOC_CHECK_=2 gdb 
> <path>/lvcreate", it may give more indication as to where the 
> actual corruption is happening.
> 
> I just recompiled the tools, and did not have any problems 
> with lvcreate, so it must be something strange.
> 
> Cheers, Andreas
> --
> Andreas Dilger  \ "If a man ate a pound of pasta and a pound 
> of antipasto,
>                  \  would they cancel out, leaving him still hungry?"
> http://www-mddsp.enel.ucalgary.ca/People/adilger/             
>   -- Dogbert
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

RE: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

Since I'm having trouble with rc-4 can I use another version that is
still compatible with everything and won't segfault!? :)

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> Sent: October 4, 2001 7:39 AM
> To: linux-lvm@sistina.com
> Subject: RE: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> 
> Ok, done...
> 
> http://www.andromedas.com/lvm/session.log
> 
> It's 35MB and I'm only on a 3MB DSL connection so it's gonna 
> take a while to load! :)
> 
> > -----Original Message-----
> > From: linux-lvm-admin@sistina.com
> > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Andreas Dilger
> > Sent: October 4, 2001 2:05 AM
> > To: linux-lvm@sistina.com
> > Subject: Re: [linux-lvm] lvcreate segfault in 
> > vg_setup_pointers_for_snapshots
> > 
> > 
> > 
> > On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> > > Wow!  Lots and LOTS of stuff!!  Didn't think it was ever going to
> > > end!!
> > 
> > Some of it is even useful.  Maybe you can include some of the
> > stuff at the end, or put it on a web page somewhere.
> > 
> > > Here's what yer looking for:
> > > 
> > > #0  lv_create (vg=0x8, lv=0xbffffb74, 
> > >     lv_name=0xbffffb98
> > > 
> > 
> "????????????????????????????????\002???!???,???<???D???T???b???s???\2
> > > 01
> > > ???\214???\227???????")
> > >     at lv_create_remove.c:42
> > > #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31
> > 
> > So it is dereferencing an uninitialized string (or so it
> > appears).  It may be that the string is corrupted elsewhere, 
> > however.  I'm surprised that
> > lv_check_name() in lv_create_remove() didn't catch this.  
> > However, this may also be a red herring, since sometimes GDB 
> > shows junk.
> > 
> > It is also suspicious because the lv and lv_name pointers are
> > 0x24 apart, which shouldn't be the case since lv_name is the 
> > first item in lv_t.
> > 
> > > #2  0x4006c177 in __libc_start_main (main=0x8049390 <strcpy+384>,
> > > argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>,
> > >     fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
> > > stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129
> > 
> > If you try running with "export MALLOC_CHECK_=2 gdb
> > <path>/lvcreate", it may give more indication as to where the 
> > actual corruption is happening.
> > 
> > I just recompiled the tools, and did not have any problems
> > with lvcreate, so it must be something strange.
> > 
> > Cheers, Andreas
> > --
> > Andreas Dilger  \ "If a man ate a pound of pasta and a pound
> > of antipasto,
> >                  \  would they cancel out, leaving him 
> still hungry?"
> > http://www-mddsp.enel.ucalgary.ca/People/adilger/             
> >   -- Dogbert
> > 
> > 
> > _______________________________________________
> > linux-lvm mailing list
> > linux-lvm@sistina.com
> > http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> > read the 
> > LVM HOW-TO at 
> > http://www.sistina.com/lvm/Pages/howto.html
> > 
> 
> 
> _______________________________________________
> 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
> 

RE: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

Hmm, 1.0.1-rc2 segfaults aswell :/  Perhaps it's something else.  I'm
using the stock compiler from 7.1.  Not too sure what it is or how to
find it.

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> Sent: October 4, 2001 10:01 AM
> To: linux-lvm@sistina.com
> Subject: RE: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> 
> Since I'm having trouble with rc-4 can I use another version 
> that is still compatible with everything and won't segfault!? :)
> 
> > -----Original Message-----
> > From: linux-lvm-admin@sistina.com
> > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> > Sent: October 4, 2001 7:39 AM
> > To: linux-lvm@sistina.com
> > Subject: RE: [linux-lvm] lvcreate segfault in 
> > vg_setup_pointers_for_snapshots
> > 
> > 
> > 
> > 
> > Ok, done...
> > 
> > http://www.andromedas.com/lvm/session.log
> > 
> > It's 35MB and I'm only on a 3MB DSL connection so it's gonna
> > take a while to load! :)
> > 
> > > -----Original Message-----
> > > From: linux-lvm-admin@sistina.com 
> > > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Andreas Dilger
> > > Sent: October 4, 2001 2:05 AM
> > > To: linux-lvm@sistina.com
> > > Subject: Re: [linux-lvm] lvcreate segfault in
> > > vg_setup_pointers_for_snapshots
> > > 
> > > 
> > > 
> > > On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> > > > Wow!  Lots and LOTS of stuff!!  Didn't think it was 
> ever going to 
> > > > end!!
> > > 
> > > Some of it is even useful.  Maybe you can include some of 
> the stuff 
> > > at the end, or put it on a web page somewhere.
> > > 
> > > > Here's what yer looking for:
> > > > 
> > > > #0  lv_create (vg=0x8, lv=0xbffffb74, 
> > > >     lv_name=0xbffffb98
> > > > 
> > > 
> > 
> "????????????????????????????????\002???!???,???<???D???T???b???s???\2
> > > > 01
> > > > ???\214???\227???????")
> > > >     at lv_create_remove.c:42
> > > > #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31
> > > 
> > > So it is dereferencing an uninitialized string (or so it 
> appears).  
> > > It may be that the string is corrupted elsewhere, however.  I'm 
> > > surprised that
> > > lv_check_name() in lv_create_remove() didn't catch this.
> > > However, this may also be a red herring, since sometimes GDB 
> > > shows junk.
> > > 
> > > It is also suspicious because the lv and lv_name pointers 
> are 0x24 
> > > apart, which shouldn't be the case since lv_name is the 
> first item 
> > > in lv_t.
> > > 
> > > > #2  0x4006c177 in __libc_start_main (main=0x8049390 
> <strcpy+384>, 
> > > > argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>,
> > > >     fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
> > > > stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129
> > > 
> > > If you try running with "export MALLOC_CHECK_=2 gdb 
> > > <path>/lvcreate", it may give more indication as to where 
> the actual 
> > > corruption is happening.
> > > 
> > > I just recompiled the tools, and did not have any problems with 
> > > lvcreate, so it must be something strange.
> > > 
> > > Cheers, Andreas
> > > --
> > > Andreas Dilger  \ "If a man ate a pound of pasta and a pound of 
> > > antipasto,
> > >                  \  would they cancel out, leaving him
> > still hungry?"
> > > http://www-mddsp.enel.ucalgary.ca/People/adilger/             
> > >   -- Dogbert
> > > 
> > > 
> > > _______________________________________________
> > > linux-lvm mailing list
> > > linux-lvm@sistina.com 
> > > http://lists.sistina.com/mailman/listinfo/linu> x-lvm read the
> > > LVM HOW-TO at 
> > > http://www.sistina.com/lvm/Pages/howto.html
> > > 
> > 
> > 
> > _______________________________________________
> > 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
> > 
> 
> 
> _______________________________________________
> linux-lvm mailing list
> linux-lvm@sistina.com 
> http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> read the 
> LVM HOW-TO at 
> http://www.sistina.com/lvm/Pages/howto.html
> 

RE: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Jason A. Lixfeld]

I managed to get the tools to compile and work with kgcc 2.91.66.

> -----Original Message-----
> From: linux-lvm-admin@sistina.com 
> [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> Sent: October 4, 2001 10:24 AM
> To: linux-lvm@sistina.com
> Subject: RE: [linux-lvm] lvcreate segfault in 
> vg_setup_pointers_for_snapshots
> 
> 
> 
> 
> Hmm, 1.0.1-rc2 segfaults aswell :/  Perhaps it's something 
> else.  I'm using the stock compiler from 7.1.  Not too sure 
> what it is or how to find it.
> 
> > -----Original Message-----
> > From: linux-lvm-admin@sistina.com
> > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> > Sent: October 4, 2001 10:01 AM
> > To: linux-lvm@sistina.com
> > Subject: RE: [linux-lvm] lvcreate segfault in 
> > vg_setup_pointers_for_snapshots
> > 
> > 
> > 
> > 
> > Since I'm having trouble with rc-4 can I use another version
> > that is still compatible with everything and won't segfault!? :)
> > 
> > > -----Original Message-----
> > > From: linux-lvm-admin@sistina.com 
> > > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Jason A. Lixfeld
> > > Sent: October 4, 2001 7:39 AM
> > > To: linux-lvm@sistina.com
> > > Subject: RE: [linux-lvm] lvcreate segfault in
> > > vg_setup_pointers_for_snapshots
> > > 
> > > 
> > > 
> > > 
> > > Ok, done...
> > > 
> > > http://www.andromedas.com/lvm/session.log
> > > 
> > > It's 35MB and I'm only on a 3MB DSL connection so it's 
> gonna take a 
> > > while to load! :)
> > > 
> > > > -----Original Message-----
> > > > From: linux-lvm-admin@sistina.com
> > > > [mailto:linux-lvm-admin@sistina.com] On Behalf Of Andreas Dilger
> > > > Sent: October 4, 2001 2:05 AM
> > > > To: linux-lvm@sistina.com
> > > > Subject: Re: [linux-lvm] lvcreate segfault in
> > > > vg_setup_pointers_for_snapshots
> > > > 
> > > > 
> > > > 
> > > > On Oct 03, 2001  23:46 -0400, Jason A. Lixfeld wrote:
> > > > > Wow!  Lots and LOTS of stuff!!  Didn't think it was
> > ever going to
> > > > > end!!
> > > > 
> > > > Some of it is even useful.  Maybe you can include some of
> > the stuff
> > > > at the end, or put it on a web page somewhere.
> > > > 
> > > > > Here's what yer looking for:
> > > > > 
> > > > > #0  lv_create (vg=0x8, lv=0xbffffb74, 
> > > > >     lv_name=0xbffffb98
> > > > > 
> > > > 
> > > 
> > 
> "????????????????????????????????\002???!???,???<???D???T???b???s???\2
> > > > > 01
> > > > > ???\214???\227???????")
> > > > >     at lv_create_remove.c:42
> > > > > #1  0x0804b03d in strcpy () at ../sysdeps/generic/strcpy.c:31
> > > > 
> > > > So it is dereferencing an uninitialized string (or so it
> > appears).
> > > > It may be that the string is corrupted elsewhere, however.  I'm
> > > > surprised that
> > > > lv_check_name() in lv_create_remove() didn't catch this.
> > > > However, this may also be a red herring, since sometimes GDB 
> > > > shows junk.
> > > > 
> > > > It is also suspicious because the lv and lv_name pointers
> > are 0x24
> > > > apart, which shouldn't be the case since lv_name is the
> > first item
> > > > in lv_t.
> > > > 
> > > > > #2  0x4006c177 in __libc_start_main (main=0x8049390
> > <strcpy+384>,
> > > > > argc=8, ubp_av=0xbffffb74, init=0x8048e88 <_init>,
> > > > >     fini=0x804b4d0 <_fini>, rtld_fini=0x4000e184 <_dl_fini>,
> > > > > stack_end=0xbffffb6c) at ../sysdeps/generic/libc-start.c:129
> > > > 
> > > > If you try running with "export MALLOC_CHECK_=2 gdb
> > > > <path>/lvcreate", it may give more indication as to where 
> > the actual
> > > > corruption is happening.
> > > > 
> > > > I just recompiled the tools, and did not have any problems with
> > > > lvcreate, so it must be something strange.
> > > > 
> > > > Cheers, Andreas
> > > > --
> > > > Andreas Dilger  \ "If a man ate a pound of pasta and a pound of
> > > > antipasto,
> > > >                  \  would they cancel out, leaving him
> > > still hungry?"
> > > > http://www-mddsp.enel.ucalgary.ca/People/adilger/             
> > > >   -- Dogbert
> > > > 
> > > > 
> > > > _______________________________________________
> > > > linux-lvm mailing list
> > > > linux-lvm@sistina.com
> > > > http://lists.sistina.com/mailman/listinfo/linu> x-lvm read the
> > > > LVM HOW-TO at 
> > > > http://www.sistina.com/lvm/Pages/howto.html
> > > > 
> > > 
> > > 
> > > _______________________________________________
> > > 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
> > > 
> > 
> > 
> > _______________________________________________
> > linux-lvm mailing list
> > linux-lvm@sistina.com
> > http://lists.sistina.com/mailman/listinfo/linu> x-lvm
> > read the 
> > LVM HOW-TO at 
> > http://www.sistina.com/lvm/Pages/howto.html
> > 
> 
> 
> _______________________________________________
> 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
> 

Re: [linux-lvm] lvcreate segfault in vg_setup_pointers_for_snapshots

[Goetz Bock]

[-- Attachment #1: Type: text/plain, Size: 368 bytes --]


On Thu, Oct 04 '01 at 11:57, Jason A. Lixfeld wrote:
> I managed to get the tools to compile and work with kgcc 2.91.66.
I did not read this thread at all, but just wanted to mention, that you
should NEVER try to compile any kernel stuff with a gcc newer than
2.95.2
-- 
Goetz Bock                                              IT Consultant
Dipl.-Inf. Univ.

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