Bug 66411 - BUG at at fs/btrfs/inode.c:1595
Summary: BUG at at fs/btrfs/inode.c:1595
Status: NEW
Alias: None
Product: File System
Classification: Unclassified
Component: btrfs (show other bugs)
Hardware: All Linux
: P1 normal
Assignee: Josef Bacik
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2013-12-03 15:52 UTC by Emil Karlson
Modified: 2014-02-20 23:22 UTC (History)
2 users (show)

See Also:
Kernel Version: 3.12.2
Tree: Mainline
Regression: No


Attachments
Full kernel log (69.18 KB, text/plain)
2013-12-03 15:52 UTC, Emil Karlson
Details
kernel log (120.37 KB, text/plain)
2014-02-20 23:22 UTC, Matt Deacalion Stevens
Details

Description Emil Karlson 2013-12-03 15:52:58 UTC
Created attachment 117261 [details]
Full kernel log

I have no idea what triggered this

mount options (rw,noatime,space_cache,autodefrag)

dmesg:

[15037.010889] ------------[ cut here ]------------
[15037.010914] WARNING: CPU: 0 PID: 31884 at fs/btrfs/ctree.c:1322 btrfs_search_old_slot+0x338/0x81d()
[15037.010919] Modules linked in: fglrx(PO) microcode
[15037.010934] CPU: 0 PID: 31884 Comm: btrfs-endio-wri Tainted: P           O 3.12.2 #1
[15037.010940] Hardware name: Gigabyte Technology Co., Ltd. GA-890GPA-UD3H/GA-890GPA-UD3H, BIOS F9 09/09/2011
[15037.010946]  0000000000000009 ffff88020e507908 ffffffff81479636 0000000000000006
[15037.010954]  0000000000000000 ffff88020e507948 ffffffff8105ecb7 0000003de23d8000
[15037.010961]  ffffffff8117e04f ffff88041d7cfa20 ffff8803270f2800 ffff88037c2fd2c0
[15037.010969] Call Trace:
[15037.010983]  [<ffffffff81479636>] dump_stack+0x46/0x58
[15037.010992]  [<ffffffff8105ecb7>] warn_slowpath_common+0x77/0x91
[15037.011002]  [<ffffffff8117e04f>] ? btrfs_search_old_slot+0x338/0x81d
[15037.011010]  [<ffffffff8105ece6>] warn_slowpath_null+0x15/0x17
[15037.011019]  [<ffffffff8117e04f>] btrfs_search_old_slot+0x338/0x81d
[15037.011031]  [<ffffffff811e17e8>] __resolve_indirect_refs+0x10f/0x481
[15037.011040]  [<ffffffff811e21dd>] find_parent_nodes+0x33b/0x5d6
[15037.011049]  [<ffffffff811e2c8b>] iterate_extent_inodes+0xc9/0x1d6
[15037.011057]  [<ffffffff8119accf>] ? record_extent_backrefs+0xc1/0xc1
[15037.011065]  [<ffffffff8119accf>] ? record_extent_backrefs+0xc1/0xc1
[15037.011073]  [<ffffffff811e2e17>] iterate_inodes_from_logical+0x7f/0x95
[15037.011081]  [<ffffffff8119ac69>] record_extent_backrefs+0x5b/0xc1
[15037.011089]  [<ffffffff811a3317>] btrfs_finish_ordered_io+0x79b/0x8e4
[15037.011099]  [<ffffffff811a3470>] finish_ordered_fn+0x10/0x12
[15037.011108]  [<ffffffff811be428>] worker_loop+0x15e/0x496
[15037.011117]  [<ffffffff811be2ca>] ? btrfs_queue_worker+0x26d/0x26d
[15037.011125]  [<ffffffff810760ea>] kthread+0xb5/0xbd
[15037.011133]  [<ffffffff81076035>] ? kthread_freezable_should_stop+0x43/0x43
[15037.011141]  [<ffffffff81482dfc>] ret_from_fork+0x7c/0xb0
[15037.011148]  [<ffffffff81076035>] ? kthread_freezable_should_stop+0x43/0x43
[15037.011154] ---[ end trace e48a6baec0f0bc3c ]---
[15037.011190] BTRFS critical (device sda2): unable to find logical 0 len 4096
[15037.011208] ------------[ cut here ]------------
[15037.011213] kernel BUG at fs/btrfs/inode.c:1595!
[15037.011218] invalid opcode: 0000 [#1] SMP 
[15037.011222] Modules linked in: fglrx(PO) microcode
[15037.011233] CPU: 0 PID: 31884 Comm: btrfs-endio-wri Tainted: P        W  O 3.12.2 #1
[15037.011238] Hardware name: Gigabyte Technology Co., Ltd. GA-890GPA-UD3H/GA-890GPA-UD3H, BIOS F9 09/09/2011
[15037.011243] task: ffff88029ffe4990 ti: ffff88020e506000 task.ti: ffff88020e506000
[15037.011248] RIP: 0010:[<ffffffff8119c532>]  [<ffffffff8119c532>] btrfs_merge_bio_hook+0x53/0x68
[15037.011258] RSP: 0018:ffff88020e507578  EFLAGS: 00010282
[15037.011263] RAX: 00000000ffffffea RBX: 0000000000001000 RCX: 0000000000000046
[15037.011268] RDX: 0000000000000006 RSI: 0000000000000046 RDI: ffff88042fc0d130
[15037.011272] RBP: ffff88020e507598 R08: 0000000000000002 R09: 00000000fffffffe
[15037.011277] R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000001000
[15037.011282] R13: ffff88020c761468 R14: 0000000000000008 R15: ffff8800cf4fc040
[15037.011288] FS:  00007f90535c5700(0000) GS:ffff88042fc00000(0000) knlGS:0000000000000000
[15037.011293] CS:  0010 DS: 0000 ES: 0000 CR0: 000000008005003b
[15037.011298] CR2: 00007f5cf4017038 CR3: 00000003d8ce4000 CR4: 00000000000007f0
[15037.011302] Stack:
[15037.011305]  0000000000000004 0000000000001000 0000000000001000 ffff88020e5077e8
[15037.011312]  ffff88020e5075f8 ffffffff811b0842 ffff88020e5075f8 0000000000000000
[15037.011318]  ffffea000815ccc0 0000002000000000 ffff88020e507648 0000000000001000
[15037.011325] Call Trace:
[15037.011334]  [<ffffffff811b0842>] submit_extent_page.isra.43+0xea/0x17e
[15037.011342]  [<ffffffff811b183c>] __do_readpage+0x4cc/0x56d
[15037.011349]  [<ffffffff811b0af3>] ? repair_io_failure+0x1dc/0x1dc
[15037.011359]  [<ffffffff81191cf0>] ? verify_parent_transid+0x146/0x146
[15037.011368]  [<ffffffff811ad922>] ? btrfs_lookup_ordered_extent+0x5d/0xb2
[15037.011376]  [<ffffffff811b1992>] __extent_read_full_page+0xb5/0xc4
[15037.011384]  [<ffffffff81191cf0>] ? verify_parent_transid+0x146/0x146
[15037.011392]  [<ffffffff811b3aae>] read_extent_buffer_pages+0x208/0x222
[15037.011401]  [<ffffffff81191cf0>] ? verify_parent_transid+0x146/0x146
[15037.011410]  [<ffffffff811934e5>] btree_read_extent_buffer_pages.constprop.123+0x61/0xf9
[15037.011419]  [<ffffffff811949b1>] read_tree_block+0x2c/0x45
[15037.011428]  [<ffffffff8117ba94>] read_block_for_search.isra.44+0x2b4/0x2fb
[15037.011436]  [<ffffffff8105ecc6>] ? warn_slowpath_common+0x86/0x91
[15037.011445]  [<ffffffff8117e1df>] btrfs_search_old_slot+0x4c8/0x81d
[15037.011452]  [<ffffffff811e17e8>] __resolve_indirect_refs+0x10f/0x481
[15037.011461]  [<ffffffff811e21dd>] find_parent_nodes+0x33b/0x5d6
[15037.011469]  [<ffffffff811e2c8b>] iterate_extent_inodes+0xc9/0x1d6
[15037.011477]  [<ffffffff8119accf>] ? record_extent_backrefs+0xc1/0xc1
[15037.011484]  [<ffffffff8119accf>] ? record_extent_backrefs+0xc1/0xc1
[15037.011492]  [<ffffffff811e2e17>] iterate_inodes_from_logical+0x7f/0x95
[15037.011500]  [<ffffffff8119ac69>] record_extent_backrefs+0x5b/0xc1
[15037.011508]  [<ffffffff811a3317>] btrfs_finish_ordered_io+0x79b/0x8e4
[15037.011517]  [<ffffffff811a3470>] finish_ordered_fn+0x10/0x12
[15037.011525]  [<ffffffff811be428>] worker_loop+0x15e/0x496
[15037.011533]  [<ffffffff811be2ca>] ? btrfs_queue_worker+0x26d/0x26d
[15037.011540]  [<ffffffff810760ea>] kthread+0xb5/0xbd
[15037.011548]  [<ffffffff81076035>] ? kthread_freezable_should_stop+0x43/0x43
[15037.011555]  [<ffffffff81482dfc>] ret_from_fork+0x7c/0xb0
[15037.011563]  [<ffffffff81076035>] ? kthread_freezable_should_stop+0x43/0x43
[15037.011567] Code: 8b 58 30 41 89 fa 48 8b b9 e8 01 00 00 48 8d 4d e8 48 c1 e2 09 45 31 c9 45 31 c0 44 89 d6 48 89 5d e8 e8 10 e0 01 00 85 c0 79 02 <0f> 0b 31 c0 4c 01 e3 48 39 5d e8 0f 92 c0 5a 59 5b 41 5c 5d c3 
[15037.011627] RIP  [<ffffffff8119c532>] btrfs_merge_bio_hook+0x53/0x68
[15037.011635]  RSP <ffff88020e507578>
[15037.011640] ---[ end trace e48a6baec0f0bc3d ]---
Comment 1 David Sterba 2013-12-03 16:31:16 UTC
skinny metadata extents are turned on

the warning does not look healthy:

1322         WARN_ON(btrfs_header_nritems(eb) > BTRFS_NODEPTRS_PER_BLOCK(root));

the crash:

1575 int btrfs_merge_bio_hook(int rw, struct page *page, unsigned long offset,
1576                          size_t size, struct bio *bio,
1577                          unsigned long bio_flags)
1578 {
1579         struct btrfs_root *root = BTRFS_I(page->mapping->host)->root;
1580         u64 logical = (u64)bio->bi_sector << 9;
1581         u64 length = 0;
1582         u64 map_length;
1583         int ret;
1584
1585         if (bio_flags & EXTENT_BIO_COMPRESSED)
1586                 return 0;
1587
1588         length = bio->bi_size;
1589         map_length = length;
1590         ret = btrfs_map_block(root->fs_info, rw, logical,
1591                               &map_length, NULL, 0);
1592         /* Will always return 0 with map_multi == NULL */
1593         BUG_ON(ret < 0);

EINVAL
Comment 2 Matt Deacalion Stevens 2014-02-20 23:22:18 UTC
Today I have had similar warnings in my kernel log. I'm running 3.12.9-2. Everything appears to be working fine though. I've attached the log.

I don't know if this is relevant, but I never realised systemd timer events aren't persistent across boots and I set up several units to perform backups, snapshots, scrub, fstrim, SMART tests, package updates etc. These were supposed to be staggered for different points in time, but they ALL ran on boot.
Comment 3 Matt Deacalion Stevens 2014-02-20 23:22:50 UTC
Created attachment 126891 [details]
kernel log

Note You need to log in before you can comment on or make changes to this bug.