[LU-130] Kernel crash on lustre 2.0 client (page fault in ll_file_read, NULL pointer dereference)

• Type: Bug
• Resolution: Fixed
• Priority: Blocker
• Fix Version/s: Lustre 2.1.0
• Affects Version/s: Lustre 2.0.0

Environment

# uname -r
2.6.32-71.14.1.el6.Bull.19.x86_64
# rpm -qa lustre\*
lustre-modules-2.0.0.1-2.6.32_71.14.1.el6.Bull.18.x86_64_Bull.petaflop.1.052.el6.x86_64
lustre-2.0.0.1-Bull.petaflop.1.052.el6.x86_64
lustre_e2fsprogs-1.41.10.sun2-Bull.petaflop.1.052.el6.20110128.x86_64

• Severity: 3
• Assignee: Johann Lombardi (Inactive)
• Reporter: Patrick Valentin (Inactive)

Description

Hello,

As suggested by Johann Lombardi, we open a Jira ticket for this issue which occurs more and more frequently at CEA site (Bull Customer).

Bellow is an extract of dmesg output and a stack trace:

=== from dmesg output:

BUG: unable to handle kernel NULL pointer dereference at 000000000000000a
IP: [<ffffffffa05f7cd7>] cl_vmpage_page+0x57/0x1e0 [obdclass]
PGD 3f7fc5067 PUD 3b3d2d067 PMD 0
Oops: 0000 1 SMP

=== backtrace:

PID: 27785 TASK: ffff8803b3e95240 CPU: 5 COMMAND: "fortcom"
#0 [ffff8803f7e63510] machine_kexec at ffffffff8102e77b
#1 [ffff8803f7e63570] crash_kexec at ffffffff810a6cb8
#2 [ffff8803f7e63640] oops_end at ffffffff8146a770
#3 [ffff8803f7e63670] no_context at ffffffff810378db
#4 [ffff8803f7e636c0] __bad_area_nosemaphore at ffffffff81037b65
#5 [ffff8803f7e63710] bad_area at ffffffff81037c8e
#6 [ffff8803f7e63740] do_page_fault at ffffffff8146c2e8
#7 [ffff8803f7e63790] page_fault at ffffffff81469ae5
[exception RIP: cl_vmpage_page+87]
RIP: ffffffffa05f7cd7 RSP: ffff8803f7e63848 RFLAGS: 00010202
RAX: 0000000000001218 RBX: 0000000000000002 RCX: ffffea001167e190
RDX: 0000000000001218 RSI: ffff8803b339aa48 RDI: ffff8803b339aa08
RBP: ffff8803f7e63898 R8: 0000000000000001 R9: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffff8803b339aa48
R13: ffff8803b339aa08 R14: ffff8803b3e32c98 R15: ffffea001167e190
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018
#8 [ffff8803f7e638a0] cl_page_find0 at ffffffffa05fa4ad
#9 [ffff8803f7e63980] cl_page_find at ffffffffa05fabc1
#10 [ffff8803f7e63990] ll_cl_init at ffffffffa096e4d0
#11 [ffff8803f7e63a60] ll_readpage at ffffffffa096e88a
#12 [ffff8803f7e63ad0] generic_file_aio_read at ffffffff810fb470
#13 [ffff8803f7e63bb0] vvp_io_read_start at ffffffffa0999efb
#14 [ffff8803f7e63c60] cl_io_start at ffffffffa0601be8
#15 [ffff8803f7e63cc0] cl_io_loop at ffffffffa0605710
#16 [ffff8803f7e63d30] ll_file_io_generic at ffffffffa0942f72
#17 [ffff8803f7e63dd0] ll_file_aio_read at ffffffffa094322c
#18 [ffff8803f7e63e60] ll_file_read at ffffffffa0949811
#19 [ffff8803f7e63ef0] vfs_read at ffffffff81158a45
#20 [ffff8803f7e63f30] sys_read at ffffffff81158b81
#21 [ffff8803f7e63f80] system_call_fastpath at ffffffff8100c172
RIP: 0000003c15ad41cd RSP: 00007fff46bcd1d8 RFLAGS: 00010202
RAX: 0000000000000000 RBX: ffffffff8100c172 RCX: 00007fff46bcd2a0
RDX: 0000000000028d8c RSI: 00002b747695d000 RDI: 0000000000000003
RBP: 0000000000428d8c R8: 0000000000428d8c R9: 0000000004f36aa0
R10: 00002b74768defa0 R11: 0000000000000293 R12: 0000000000028d8c
R13: 0000000000400000 R14: 0000000004f369c0 R15: 0000000000400000
ORIG_RAX: 0000000000000000 CS: 0033 SS: 002b