[LU-4493] NRS ORR crash

• Type: Bug
• Resolution: Cannot Reproduce
• Priority: Minor
• Affects Version/s: Lustre 2.4.2
• Environment: Lustre 2.4.2 server, was doing some streaming writes when policy was enabled
• Severity: 3
• Assignee: WC Triage
• Reporter: Andreas Dilger

Description

kernel BUG at mm/slab.c:2835!
invalid opcode: 0000 [#1] 
SMP
last sysfs file: /sys/devices/pci0000:00/0000:00:1e.0/0000:01:09.0/host5/target5:0:0/5:0:0:0/block/sdf/dev
CPU 0
Pid: 24075, comm: ll_ost_io00_001 Not tainted .6.32-358.18.1.el6_lustre.g63f298d.x86_64 #1
RIP: 0010:[<ffffffff81167443>] 
 [<ffffffff81167443>] cache_grow+0x313/0x320
RSP: 0018:ffff880036e51c10  EFLAGS: 00010002
RAX: ffff880055516c00 RBX: ffff88006bcb4000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000041212 RDI: ffff88006bcb4000
RBP: ffff880036e51c70 R08: 0000000000000000 R09: 0000000000000532
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000041212
R13: ffff880055516bc0 R14: 0000000000000010 R15: 0000000000000000
FS:  00007f3761dc5700(0000) GS:ffff880002200000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 00007f06f4801000 CR3: 000000000ea0c000 CR4: 00000000000007f0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process ll_ost_io00_001 (pid: 24075, threadinfo ffff880036e50000, task ffff8800378aaaa0)
[<ffffffff81167652>] cache_alloc_refill+0x202/0x240
[<ffffffff8116711e>] kmem_cache_alloc_node+0x1be/0x1d0
[<ffffffffa036b8a1>] cfs_mem_cache_cpt_alloc+0x41/0x50 [libcfs]
[<ffffffffa06bd7b9>] nrs_orr_res_get+0x5d9/0xb90 [ptlrpc]
[<ffffffffa06b3b56>] nrs_resource_get+0x56/0x110 [ptlrpc]
[<ffffffffa0674850>] ? lustre_swab_niobuf_remote+0x0/0x30 [ptlrpc]
[<ffffffffa06b451b>] nrs_resource_get_safe+0x8b/0x100 [ptlrpc]
[<ffffffffa06b6a98>] ptlrpc_nrs_req_initialize+0x38/0x90 [ptlrpc]
[<ffffffffa0683df0>] ptlrpc_main+0x1180/0x1700 [ptlrpc]

Issue Links

• is related to: LU-398 NRS (Network Request Scheduler )