Actions
Bug #17070
closedKernel complains about "Bad page state" when cephfs tries to free page
Status:
Resolved
Priority:
Normal
Assignee:
-
Category:
fs/ceph
Target version:
-
% Done:
0%
Source:
Community (dev)
Tags:
Backport:
Regression:
No
Severity:
3 - minor
Reviewed:
Affected Versions:
ceph-qa-suite:
Crash signature (v1):
Crash signature (v2):
Description
This issue didn't cause cephfs getting hang or crashing directly, but it caused upper application process (here is classification process) getting hang and then the whole system became unresponsive. From our observer, this issue may happen when there is very low free memory.
Linux kernel version: 3.10.94
Aug 15 22:42:52 kernel: BUG: Bad page state: 90 messages suppressed Aug 15 22:42:52 kernel: BUG: Bad page state in process classification. pfn:c35aab Aug 15 22:42:52 kernel: page:ffffea0030d6aac0 count:0 mapcount:0 mapping: (null) index:0x172 Aug 15 22:42:52 kernel: page flags: 0x2ffc0000000020(lru) Aug 15 22:42:52 kernel: CPU: 6 PID: 12752 Comm: classification. Tainted: P B O 3.10.94-1-tlinux2-0036.tl2 #1 Aug 15 22:42:52 kernel: Hardware name: HP ProLiant SL270s Gen8 SE/, BIOS P75 11/02/2014 Aug 15 22:42:52 kernel: 000ffc0000000000 ffff880f12ecd9d8 ffffffff81ad48d8 ffff880f12ecd9f0 Aug 15 22:42:52 kernel: ffffffff81acdd15 0000000000000000 ffff880f12ecda30 ffffffff81108928 Aug 15 22:42:52 kernel: ffffea0030d6aac0 ffffea0030d6aac0 002ffc0000000020 0000000000000000 Aug 15 22:42:52 kernel: Call Trace: Aug 15 22:42:52 kernel: [<ffffffff81ad48d8>] dump_stack+0x19/0x1b Aug 15 22:42:52 kernel: [<ffffffff81acdd15>] bad_page.part.59+0xcf/0xe8 Aug 15 22:42:52 kernel: [<ffffffff81108928>] free_pages_prepare+0x148/0x160 Aug 15 22:42:52 kernel: [<ffffffff811091b1>] free_hot_cold_page+0x31/0x130 Aug 15 22:42:52 kernel: [<ffffffff81109397>] __free_pages+0x47/0x50 Aug 15 22:42:52 kernel: [<ffffffffa0bd57de>] ceph_release_page_vector+0x2e/0x50 [libceph] Aug 15 22:42:52 kernel: [<ffffffffa0c01d36>] start_read+0x266/0x400 [ceph] Aug 15 22:42:52 kernel: [<ffffffffa0c01f36>] ceph_readpages+0x66/0xd0 [ceph] Aug 15 22:42:52 kernel: [<ffffffff8110ca9e>] __do_page_cache_readahead+0x1ae/0x240 Aug 15 22:42:52 kernel: [<ffffffff8110d191>] ra_submit+0x21/0x30 Aug 15 22:42:52 kernel: [<ffffffff81103a7c>] filemap_fault+0x35c/0x430 Aug 15 22:42:52 kernel: [<ffffffff81124352>] __do_fault+0x72/0x4f0 Aug 15 22:42:52 kernel: [<ffffffff8112732e>] handle_pte_fault+0x23e/0x990 Aug 15 22:42:52 kernel: [<ffffffff811287dd>] handle_mm_fault+0x31d/0x6e0 Aug 15 22:42:52 kernel: [<ffffffff81ae1961>] __do_page_fault+0x151/0x530 Aug 15 22:42:52 kernel: [<ffffffff81421c64>] ? blk_finish_plug+0x14/0x40 Aug 15 22:42:52 kernel: [<ffffffff81123b35>] ? SyS_madvise+0x535/0x780 Aug 15 22:42:52 kernel: [<ffffffff81ae1d4e>] do_page_fault+0xe/0x10 Aug 15 22:42:52 kernel: [<ffffffff81ade5e2>] page_fault+0x22/0x30
Actions