Yes, there was a bug in 1.6.5.x. Should be fixed in 1.6.6. See
https://bugzilla.lustre.org/show_bug.cgi?id=14949
Kevin
subbu kl wrote:> I am trying to run some sanity performance tests on my test Lustre
> setup with 1 OSS adn 1 MDS/MGS.
> obdfilter-survey always cuses kernel panic where it runs. Am I missing
> something?
>
> Any help greatly appreciated.
>
> details as follows -
>
> [root at p209 ~]# uname -a
> Linux p209 2.6.18-53.1.14.el5_lustre.1.6.5.1smp #1 SMP Thu Jun 26
> 01:38:50 EDT 2008 i686 athlon i386 GNU/Linux
> [root at p209 ~]# lctl dl
> 6 UP mgc MGC172.24.198.31 at tcp b34f89bb-1a8f-ef5a-ee91-07c17f08ea85 5
> 7 UP ost OSS OSS_uuid 3
> 8 UP obdfilter lustre-OST0000 lustre-OST0000_UUID 5
> [root at p209 ~]# nobjhi=2 thrhi=2 size=1024 case=disk sh obdfilter-survey
> BUG: unable to handle kernel NULL pointer dereference at virtual
> address 0000004
> printing eip:
> dd29fc5d
> *pde = 00000000
> Oops: 0000 [#1]
> SMP
> last sysfs file: /block/hda/removable
> Modules linked in: obdecho(U) obdfilter(U) ost(U) osc(U) loop(U)
> autofs4(U) hid)
> CPU: 0
> EIP: 0060:[<dd29fc5d>] Not tainted VLI
> EFLAGS: 00010246 (2.6.18-53.1.14.el5_lustre.1.6.5.1smp #1)
> EIP is at filter_connect+0x18d/0xe50 [obdfilter]
> eax: 00000000 ebx: cdf07d94 ecx: 00000000 edx: 0000031c
> esi: 00000000 edi: 00000000 ebp: cd890cf8 esp: cdf07cf8
> ds: 007b es: 007b ss: 0068
> Process lctl (pid: 4481, ti=cdf06000 task=dbe65aa0 task.ti=cdf06000)
> Stack: 00000000 3033057d ce636630 00f07d68 0000000a ce7b2000 ffffffff
> 00000000
> dcb25c72 0000057d dcb25533 dcb1e10e 0000004a 00000000 d0267000
> 00000010
> 00000048 d91d2600 cd890cf8 dd29fad0 cffcfd60 dd0425ea cd1e57c0
> 00000000
> Call Trace:
> [<dd29fad0>] filter_connect+0x0/0xe50 [obdfilter]
> [<dd0425ea>] echo_client_setup+0xcaa/0x16d0 [obdecho]
> [<dcaf1562>] class_setup+0x482/0x1710 [obdclass]
> [<dcaf5dfe>] class_process_config+0x1e9e/0x3120 [obdclass]
> [<dcac3378>] class_handle_ioctl+0x3d98/0x4140 [obdclass]
> [<dca9c015>] obd_class_open+0x15/0x20 [obdclass]
> [<c052f478>] misc_open+0x115/0x1c5
> [<c045482f>] find_get_page+0x18/0x38
> [<c04572e3>] filemap_nopage+0x192/0x30b
> [<dcac3720>] obd_class_ioctl+0x0/0x10 [obdclass]
> [<dca9c101>] obd_class_ioctl+0x81/0x440 [obdclass]
> [<c047f2d7>] do_ioctl+0x47/0x5d
> [<c047f537>] vfs_ioctl+0x24a/0x25c
> [<c047f591>] sys_ioctl+0x48/0x5f
> [<c0404ed3>] syscall_call+0x7/0xb
> ======================> Code: c0 0f b6 c0 85 c0 89 44 24 34 0f 85 96 0b
00 00 8b 54 24 58 8b
> 44 24 38 e
> EIP: [<dd29fc5d>] filter_connect+0x18d/0xe50 [obdfilter] SS:ESP
> 0068:cdf07cf8
> <0>Kernel panic - not syncing: Fatal exception
>
> ~subbu
> ------------------------------------------------------------------------
>
> _______________________________________________
> Lustre-discuss mailing list
> Lustre-discuss at lists.lustre.org
> http://lists.lustre.org/mailman/listinfo/lustre-discuss
>