[LU-12830] RHEL8.3 and ZFS: oom on OSS

• Type: Bug
• Resolution: Cannot Reproduce
• Priority: Minor
• Affects Version/s: Lustre 2.14.0, Lustre 2.15.0
• Labels: ZFS
• Severity: 3
• Assignee: WC Triage
• Reporter: Maloo

Description

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

This issue relates to the following test suite run: https://testing.whamcloud.com/test_sets/1e6f3bc6-e5ef-11e9-b62b-52540065bddc

test_bonnie failed with oom on OSS:

[16526.881544] Lustre: DEBUG MARKER: == sanity-benchmark test bonnie: bonnie++ ============================================================ 14:37:57 (1570027077)
[16528.099983] Lustre: DEBUG MARKER: /usr/sbin/lctl mark min OST has 10511360kB available, using 3438712kB file size
[16528.357585] Lustre: DEBUG MARKER: min OST has 10511360kB available, using 3438712kB file size
[16567.214746] irqbalance invoked oom-killer: gfp_mask=0x200da, order=0, oom_score_adj=0
[16567.215741] irqbalance cpuset=/ mems_allowed=0
[16567.216221] CPU: 1 PID: 1179 Comm: irqbalance Kdump: loaded Tainted: P           OE  ------------   3.10.0-957.27.2.el7_lustre.x86_64 #1
[16567.217451] Hardware name: Red Hat KVM, BIOS 0.5.1 01/01/2011
[16567.218033] Call Trace:
[16567.218336]  [<ffffffffad565147>] dump_stack+0x19/0x1b
[16567.218874]  [<ffffffffad55fb6a>] dump_header+0x90/0x229
[16567.219422]  [<ffffffffad572b1f>] ? notifier_call_chain+0x4f/0x70
[16567.220055]  [<ffffffffacec91c8>] ? __blocking_notifier_call_chain+0x58/0x70
[16567.220779]  [<ffffffffacfbbaae>] check_panic_on_oom+0x2e/0x60
[16567.221379]  [<ffffffffacfbbecb>] out_of_memory+0x23b/0x4f0
[16567.221938]  [<ffffffffad56066e>] __alloc_pages_slowpath+0x5d6/0x724
[16567.222585]  [<ffffffffacfc2524>] __alloc_pages_nodemask+0x404/0x420
[16567.223225]  [<ffffffffad0128c5>] alloc_pages_vma+0xb5/0x200
[16567.223840]  [<ffffffffad000b15>] __read_swap_cache_async+0x115/0x190
[16567.224491]  [<ffffffffad000bb6>] read_swap_cache_async+0x26/0x60
[16567.225104]  [<ffffffffad000c9c>] swapin_readahead+0xac/0x110
[16567.225690]  [<ffffffffacfead92>] handle_pte_fault+0x812/0xd10
[16567.226280]  [<fffffffface2a621>] ? __switch_to+0x151/0x580
[16567.226858]  [<ffffffffacfed3ad>] handle_mm_fault+0x39d/0x9b0
[16567.227444]  [<ffffffffacec6efd>] ? hrtimer_start_range_ns+0x1ed/0x3c0
[16567.228100]  [<ffffffffad572603>] __do_page_fault+0x203/0x4f0
[16567.228685]  [<ffffffffad5729d6>] trace_do_page_fault+0x56/0x150
[16567.229287]  [<ffffffffad571f62>] do_async_page_fault+0x22/0xf0
[16567.229890]  [<ffffffffad56e798>] async_page_fault+0x28/0x30

VVVVVVV DO NOT REMOVE LINES BELOW, Added by Maloo for auto-association VVVVVVV
ost-pools test_23b - trevis-21vm3 crashed during ost-pools test_23b

Issue Links

• is duplicated by: LU-17924 obdfilter-survey test_1a: OSS hit OOM
• is duplicated by: LU-14643 ost-pools test_23b: 'trevis-57vm8 crashed during ost-pools test_23b
• is duplicated by: LU-14812 ost-pools test_23b OOM crash
• is related to: LU-13162 parallel-scale test_statahead: mdsrate invoked oom-killer
• is related to: LU-13594 register OOM callback in Lustre
• is related to: LU-12864 sanity-benchmark test_iozone crashes with OOM on clients
• is related to: LU-10687 sanity-benchmark test iozone hangs with client OOM
• is related to: LU-14456 ost-pools test_23b: ll_ost_io00_007 invoked oom-killer
• is related to: LU-12336 Update ZFS Version to 0.8.2
• is related to: LU-13178 Update ZFS Version to 0.8.3