[LU-3598] Failure on test suite sanity-benchmark test_iozone: page allocation failure

• Type: Bug
• Resolution: Duplicate
• Priority: Minor
• Affects Version/s: Lustre 2.5.0, Lustre 2.5.1
• Environment: server and client: lustre-master build #1560
• Severity: 3
• Assignee: WC Triage
• Reporter: Maloo

Description

This issue was created by maloo for sarah <[e-mail]>

This issue relates to the following test suite run: http://maloo.whamcloud.com/test_sets/02803404-ed9b-11e2-8e3a-52540035b04c.

The sub-test test_iozone failed with the following error:

test failed to respond and timed out

client console:

12:52:41:Lustre: DEBUG MARKER: == sanity-benchmark test iozone: iozone == 12:52:37 (1373831557)
12:52:41:Lustre: DEBUG MARKER: /usr/sbin/lctl mark min OST has 1850592kB available, using 3844936kB file size
12:52:41:Lustre: DEBUG MARKER: min OST has 1850592kB available, using 3844936kB file size
12:54:03:ptlrpcd_1: page allocation failure. order:1, mode:0x40
12:54:03:Pid: 28900, comm: ptlrpcd_1 Not tainted 2.6.32-358.11.1.el6.x86_64 #1
12:54:03:Call Trace:
12:54:03: [<ffffffff8112c157>] ? __alloc_pages_nodemask+0x757/0x8d0
12:54:03: [<ffffffffa06627df>] ? ptlrpc_set_add_new_req+0xcf/0x150 [ptlrpc]
12:54:03: [<ffffffff81166b02>] ? kmem_getpages+0x62/0x170
12:54:03: [<ffffffff8116771a>] ? fallback_alloc+0x1ba/0x270
12:54:03: [<ffffffff8116716f>] ? cache_grow+0x2cf/0x320
12:54:03: [<ffffffff81167499>] ? ____cache_alloc_node+0x99/0x160
12:54:03: [<ffffffffa048fd57>] ? LNetMDAttach+0x157/0x5a0 [lnet]
12:54:03: [<ffffffff81168269>] ? __kmalloc+0x189/0x220
12:54:03: [<ffffffffa048fd57>] ? LNetMDAttach+0x157/0x5a0 [lnet]
12:54:03: [<ffffffffa0670b55>] ? ptlrpc_register_bulk+0x265/0x9d0 [ptlrpc]
12:54:03: [<ffffffffa069ea48>] ? sptlrpc_import_sec_ref+0x28/0x50 [ptlrpc]
12:54:03: [<ffffffffa0671fa2>] ? ptl_send_rpc+0x232/0xc40 [ptlrpc]
12:54:03: [<ffffffffa06a65b4>] ? sptlrpc_req_refresh_ctx+0x164/0x970 [ptlrpc]
12:54:03: [<ffffffffa0890971>] ? osc_io_unplug+0x11/0x20 [osc]
12:54:03: [<ffffffffa06676bb>] ? ptlrpc_send_new_req+0x45b/0x7a0 [ptlrpc]
12:54:03: [<ffffffff81055ab3>] ? __wake_up+0x53/0x70
12:54:03: [<ffffffffa066b3a8>] ? ptlrpc_check_set+0x878/0x1b20 [ptlrpc]
12:54:03: [<ffffffffa06976cb>] ? ptlrpcd_check+0x53b/0x560 [ptlrpc]
12:54:03: [<ffffffff8109705c>] ? remove_wait_queue+0x3c/0x50
12:54:03: [<ffffffffa0697b50>] ? ptlrpcd+0x190/0x380 [ptlrpc]
12:54:03: [<ffffffff81063310>] ? default_wake_function+0x0/0x20
12:54:03: [<ffffffffa06979c0>] ? ptlrpcd+0x0/0x380 [ptlrpc]
12:54:03: [<ffffffff81096936>] ? kthread+0x96/0xa0
12:54:03: [<ffffffff8100c0ca>] ? child_rip+0xa/0x20
12:54:03: [<ffffffff810968a0>] ? kthread+0x0/0xa0
12:54:03: [<ffffffff8100c0c0>] ? child_rip+0x0/0x20
12:54:03:Mem-Info:
12:54:03:Node 0 DMA per-cpu:
12:54:03:CPU    0: hi:    0, btch:   1 usd:   0
12:54:03:Node 0 DMA32 per-cpu:
12:54:03:CPU    0: hi:  186, btch:  31 usd: 159

Issue Links

• duplicates: LU-4357 page allocation failure. mode:0x40 caused by missing __GFP_WAIT flag
• is related to: LU-3585 Client panic during IOR single file per process: Lnet out of Memory