public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* Re: PROBLEM: block loop device hangs
  2001-03-03 15:06 PROBLEM: block loop device hangs Steven Brooks
@ 2001-03-03 13:35 ` Arnaldo Carvalho de Melo
  2001-03-03 15:10 ` Jens Axboe
  1 sibling, 0 replies; 3+ messages in thread
From: Arnaldo Carvalho de Melo @ 2001-03-03 13:35 UTC (permalink / raw)
  To: Steven Brooks; +Cc: linux-kernel

Em Sat, Mar 03, 2001 at 09:06:30AM -0600, Steven Brooks escreveu:
> When mounting a file using the loopback device, the mount program hangs
> for ever.  Other than that, the system is still usable.
> 
> Dist: Redhat-7
> Kernel: 2.4.2 (compiled with kgcc, i.e. egcs-2.91.66)

FAQ, try 2.4.2-ac10, several fixes to looback device

- Arnaldo

^ permalink raw reply	[flat|nested] 3+ messages in thread

* PROBLEM: block loop device hangs
@ 2001-03-03 15:06 Steven Brooks
  2001-03-03 13:35 ` Arnaldo Carvalho de Melo
  2001-03-03 15:10 ` Jens Axboe
  0 siblings, 2 replies; 3+ messages in thread
From: Steven Brooks @ 2001-03-03 15:06 UTC (permalink / raw)
  To: linux-kernel

When mounting a file using the loopback device, the mount program hangs
for ever.  Other than that, the system is still usable.

Dist: Redhat-7
Kernel: 2.4.2 (compiled with kgcc, i.e. egcs-2.91.66)
CPU: Pentium II 400
Mem: 128MB RAM + 512MB swap
(More information below)

For me, it is 100% reproducable, no matter what I try to mount:
	1. an ISO image from a Win2000 FAT32 partition, from hdb
(IDE/ATAPI drive)
	2. an ext2 image from an ext2 partition, from hdb (IDE/ATAPI
drive)
	3. an ext2 image from a Win2000 FAT32 partition, from hda
(a different IDE/ATAPI drive)

The simplest way for me to reproduce it is:
	dd if=/dev/zero of=t bs=1024k count=1
	mke2fs t
	mount -o loop t /mnt/cdrom

Mount then hangs, and never recovers.  I cannot even kill it using
kill -9.

An strace of mount shows (they all die in the exact same place):
	[...]
	open("t", O_RDWR)                       = 3
	open("/dev/loop0", O_RDWR)              = 4
	mlockall(0x3, 0xbffff848)               = 0
	ioctl(4, LOOP_SET_FD, 0x3)              = 0
	ioctl(4, LOOP_SET_STATUS, 0xbffff828)   = 0
	close(4)                                = 0
	close(3)                                = 0
	rt_sigprocmask(SIG_BLOCK, ~[TRAP SEGV], NULL, 8) = 0
	stat64("/dev/loop0", {st_mode=S_IFBLK|0660, st_rdev=makedev(7,
	0), ...}) = 0
	open("/dev/loop0", O_RDONLY)            = 3
	lseek(3, 1024, SEEK_SET)                = 1024
	read(3,

(The last line is left unfinished by strace.)

ps auwx shows:
	[...]
	root       818  0.0  1.2  1564 1548 pts/0    DL   08:44   0:00 mount -o 
loop t /mnt/cdrom
	[...]

I have lots of memory available; free shows:
	             total    used    free  shared    buffers   cached
	Mem:        127156  124884    2272       0      19140    49148
	-/+ buffers/cache:   56596   70560
	Swap:       530104      60  530044

Other than that, the system is still fine.  I can umount and mount
other, real block devices (attempting to mount another loopback just
causes another process to hang).

I can't keep up with the kernel traffic.  If you need any more
information, please e-mail me directly.

Thanks,
Steven

--------

/proc/version:
Linux version 2.4.2 (root@minneyar) (gcc version egcs-2.91.66
19990314/Linux (egcs-1.1.2 release)) #5 Fri Mar 2 08:09:35 CST 2001

/proc/cpuinfo:
processor       : 0
vendor_id       : GenuineIntel
cpu family      : 6
model           : 5
model name      : Pentium II (Deschutes)
stepping        : 3
cpu MHz         : 399.327
cache size      : 512 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 2
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca 
cmov pat pse36 mmx fxsr
bogomips        : 796.26

/proc/modules:
loop                    7584   2 (autoclean)
tdfx                   53104   1
nfsd                   70992   8 (autoclean)
lockd                  49744   1 (autoclean) [nfsd]
sunrpc                 59136   1 (autoclean) [nfsd lockd]
8139too                15104   1 (autoclean)
agpgart                13952   1
nls_iso8859-1           2832   2 (autoclean)
nls_cp437               4352   2 (autoclean)
vfat                   11504   2 (autoclean)
fat                    31744   0 (autoclean) [vfat]
es1371                 26832   0
soundcore               3856   4 [es1371]
ac97_codec              7920   0 [es1371]
usb-uhci               22544   0 (unused)
usbcore                47408   1 [usb-uhci]



^ permalink raw reply	[flat|nested] 3+ messages in thread

* Re: PROBLEM: block loop device hangs
  2001-03-03 15:06 PROBLEM: block loop device hangs Steven Brooks
  2001-03-03 13:35 ` Arnaldo Carvalho de Melo
@ 2001-03-03 15:10 ` Jens Axboe
  1 sibling, 0 replies; 3+ messages in thread
From: Jens Axboe @ 2001-03-03 15:10 UTC (permalink / raw)
  To: Steven Brooks; +Cc: linux-kernel

On Sat, Mar 03 2001, Steven Brooks wrote:
> When mounting a file using the loopback device, the mount program hangs
> for ever.  Other than that, the system is still usable.

Please read lkml archives before posting, this problem has been
all over the list for the past two weeks.

Use the latest 2.4.2-ac or wait for 2.4.3-pre2

-- 
Jens Axboe


^ permalink raw reply	[flat|nested] 3+ messages in thread

end of thread, other threads:[~2001-03-03 15:14 UTC | newest]

Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-03-03 15:06 PROBLEM: block loop device hangs Steven Brooks
2001-03-03 13:35 ` Arnaldo Carvalho de Melo
2001-03-03 15:10 ` Jens Axboe

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox