* [linux-lvm] LVM on IA64
@ 2001-10-22 6:00 Stefan Fent
2001-10-22 7:29 ` Patrick Caulfield
0 siblings, 1 reply; 3+ messages in thread
From: Stefan Fent @ 2001-10-22 6:00 UTC (permalink / raw)
To: Linux-LVM
Hi,
On IA64 lvcreate segfaults when 'lv_allocated_le'
is bigger than 1 (mostly 125).
I need information about where this value is created, and what it is good
for, as this seem to happen randomly.
It segfaults in vg_cfgrestore() at:
if ( ( vg->lv[l]->lv_current_pe =
malloc ( vg->lv[l]->lv_allocated_le * \
sizeof ( pe_t))) == NULL) {
where vg->lv[l]->lv_allocated_le==125
gdb backtrace:
Program received signal SIGSEGV, Segmentation fault.
0x20000000001a0fd1 in chunk_alloc () from /lib/libc.so.6.1
(gdb) bt
#0 0x20000000001a0fd1 in chunk_alloc () from /lib/libc.so.6.1
#1 0x200000000019f720 in malloc () from /lib/libc.so.6.1
#2 0x200000000007af10 in vg_cfgrestore (vg_name=0x80000fffffffb9d4
"volg02",
vg_backup_path=0x80000fffffffa870 "/etc/lvmtab.d/volg02", opt_v=0,
vg=0x20000000000aab30)
at vg_cfgrestore.c:177
#3 0x2000000000067540 in lvm_tab_vg_read_with_pv_and_lv
(vg_name=0x80000fffffffb9d4 "volg02",
vg=0x80000fffffffae38) at lvm_tab_vg_read_with_pv_and_lv.c:60
#4 0x4000000000008420 in main (argc=3, argv=0x80000fffffffb728) at
libio.h:633
(gdb)
any help appreciated.
Kernel is 2.4.7 with LVM-drivers 0.9.1_beta7
lvm is 1.0.1-rc2
Mit freundlichen Grüssen / With kind regards,
Stefan Fent
--
SuSE GmbH, Tel: +49-911-7405331
Schanzaeckerstr. 10, Fax: +49-911-3206727
90443 Nuernberg, Email: sf@suse.de
Germany WWW: http://www.suse.de
warning: declaration does not declare anything
^ permalink raw reply [flat|nested] 3+ messages in thread* Re: [linux-lvm] LVM on IA64
2001-10-22 6:00 [linux-lvm] LVM on IA64 Stefan Fent
@ 2001-10-22 7:29 ` Patrick Caulfield
2001-10-24 5:29 ` Patrick Caulfield
0 siblings, 1 reply; 3+ messages in thread
From: Patrick Caulfield @ 2001-10-22 7:29 UTC (permalink / raw)
To: linux-lvm
On Mon, Oct 22, 2001 at 01:00:56PM +0200, Stefan Fent wrote:
> Hi,
>
> On IA64 lvcreate segfaults when 'lv_allocated_le'
> is bigger than 1 (mostly 125).
> I need information about where this value is created, and what it is good
> for, as this seem to happen randomly.
> Program received signal SIGSEGV, Segmentation fault.
> 0x20000000001a0fd1 in chunk_alloc () from /lib/libc.so.6.1
> (gdb) bt
> #0 0x20000000001a0fd1 in chunk_alloc () from /lib/libc.so.6.1
> #1 0x200000000019f720 in malloc () from /lib/libc.so.6.1
> #2 0x200000000007af10 in vg_cfgrestore (vg_name=0x80000fffffffb9d4
> "volg02",
> vg_backup_path=0x80000fffffffa870 "/etc/lvmtab.d/volg02", opt_v=0,
> vg=0x20000000000aab30)
> at vg_cfgrestore.c:177
> #3 0x2000000000067540 in lvm_tab_vg_read_with_pv_and_lv
> (vg_name=0x80000fffffffb9d4 "volg02",
> vg=0x80000fffffffae38) at lvm_tab_vg_read_with_pv_and_lv.c:60
> #4 0x4000000000008420 in main (argc=3, argv=0x80000fffffffb728) at
> libio.h:633
> (gdb)
If it's crashing in malloc then almost certainly some memory is being
overwriten elsewhere in the code.
I know rc4 works fine on Alpha so that should rule out any 64bit related
problems. It's probably worth trying ElectricFence or some such malloc debugger
to isolate just where the corruption is occurring.
patrick
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2001-10-24 5:29 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-10-22 6:00 [linux-lvm] LVM on IA64 Stefan Fent
2001-10-22 7:29 ` Patrick Caulfield
2001-10-24 5:29 ` Patrick Caulfield
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox