Bug 903794

Summary:	[abrt]: WARNING: at fs/btrfs/extent-tree.c:6297 btrfs_alloc_free_block+0x361/0x370 [btrfs]()
Product:	[Fedora] Fedora	Reporter:	Andrew McNabb <amcnabb>
Component:	kernel	Assignee:	Josef Bacik <josef>
Status:	CLOSED UPSTREAM	QA Contact:	Fedora Extras Quality Assurance <extras-qa>
Severity:	unspecified	Docs Contact:
Priority:	unspecified
Version:	18	CC:	gansalmon, itamar, jonathan, josef, kernel-maint, madhu.chinakonda
Target Milestone:	---	Keywords:	Triaged
Target Release:	---
Hardware:	x86_64
OS:	Unspecified
Whiteboard:	abrt_hash:598c6d3c6d754c5c40252e1653fae92d86090420
Fixed In Version:		Doc Type:	Bug Fix
Doc Text:		Story Points:	---
Clone Of:		Environment:
Last Closed:	2013-09-23 20:45:32 UTC	Type:	---
Regression:	---	Mount Type:	---
Documentation:	---	CRM:
Verified Versions:		Category:	---
oVirt Team:	---	RHEL 7.3 requirements from Atomic Host:
Cloudforms Team:	---	Target Upstream Version:
Embargoed:

Description Andrew McNabb 2013-01-24 20:42:09 UTC

Description of problem:


Additional info:
WARNING: at fs/btrfs/extent-tree.c:6297 btrfs_alloc_free_block+0x361/0x370 [btrfs]()
Hardware name: MS-7596
Modules linked in: iptable_raw xt_CT nf_conntrack bridge stp llc f71882fg hwmon_vid snd_hda_codec_realtek snd_hda_intel snd_hda_codec snd_hwdep snd_seq snd_seq_device snd_pcm edac_core sp5100_tco i2c_piix4 snd_page_alloc snd_timer snd edac_mce_amd soundcore k10temp vhost_net tun macvtap macvlan kvm_amd shpchp ppdev kvm serio_raw parport_pc parport microcode nfsd auth_rpcgss uinput nfs_acl lockd binfmt_misc btrfs ata_generic libcrc32c radeon pata_acpi zlib_deflate i2c_algo_bit drm_kms_helper r8169 ttm pata_atiixp drm mii wmi i2c_core sunrpc
Pid: 2135, comm: btrfs Not tainted 3.7.2-204.fc18.x86_64 #1
Call Trace:
 [<ffffffff8105e67f>] warn_slowpath_common+0x7f/0xc0
 [<ffffffff8105e6da>] warn_slowpath_null+0x1a/0x20
 [<ffffffffa0213d41>] btrfs_alloc_free_block+0x361/0x370 [btrfs]
 [<ffffffffa0200d95>] ? __btrfs_cow_block+0x325/0x510 [btrfs]
 [<ffffffffa0242a33>] ? read_extent_buffer+0xc3/0x120 [btrfs]
 [<ffffffffa0200b9a>] __btrfs_cow_block+0x12a/0x510 [btrfs]
 [<ffffffffa023c2d5>] ? mark_extent_buffer_accessed+0x55/0x70 [btrfs]
 [<ffffffffa02010f7>] btrfs_cow_block+0xf7/0x200 [btrfs]
 [<ffffffffa0267137>] do_relocation+0x477/0x540 [btrfs]
 [<ffffffffa026ac76>] relocate_tree_blocks+0x676/0x6b0 [btrfs]
 [<ffffffffa0269448>] ? add_data_references+0x258/0x290 [btrfs]
 [<ffffffffa026bc1d>] relocate_block_group+0x42d/0x6a0 [btrfs]
 [<ffffffffa026c03d>] btrfs_relocate_block_group+0x1ad/0x2e0 [btrfs]
 [<ffffffffa0244ce5>] btrfs_relocate_chunk.isra.53+0x75/0x730 [btrfs]
 [<ffffffffa0241897>] ? free_extent_buffer+0x37/0x90 [btrfs]
 [<ffffffffa0249377>] btrfs_balance+0x887/0xe10 [btrfs]
 [<ffffffffa024f423>] btrfs_ioctl_balance+0x123/0x430 [btrfs]
 [<ffffffffa0253939>] btrfs_ioctl+0xe59/0x1970 [btrfs]
 [<ffffffff8115ab91>] ? handle_mm_fault+0x261/0x330
 [<ffffffff816395f4>] ? __do_page_fault+0x244/0x4e0
 [<ffffffff8115ebf0>] ? __vma_link_rb+0x30/0x40
 [<ffffffff811a61d9>] do_vfs_ioctl+0x99/0x580
 [<ffffffff8116060f>] ? do_brk+0x1ff/0x370
 [<ffffffff811a6751>] sys_ioctl+0x91/0xb0
 [<ffffffff8163ded9>] system_call_fastpath+0x16/0x1b

Comment 1 Andrew McNabb 2013-01-24 20:45:40 UTC

This occurred during a rebalance from a single disk to raid10 across four disks:

btrfs balance start -dconvert=raid10 -mconvert=raid10 /aml

Since then, I've seen hundreds of additional kernel oopses which may have been caused by this (abrt-cli refuses to report them because the kernel is now tainted).

Comment 2 Andrew McNabb 2013-01-24 20:55:42 UTC

Hmm... it looks like the hundreds of kernel warnings are all identical to this one. They are occurring very frequently; they seem to come in pairs within a fraction of a second separated from other pairs by about 5 seconds: 5505.441974, 5505.609500, 5512.104774, 5512.273001, 5517.795071, 5517.963667, 5523.472310, 5523.638308, 5528.945711, 5529.114985, 5534.269554, 5534.392667. The btrfs balance seems to be continuing at a dog-slow pace.

I am using kernel-3.7.2-204.fc18.x86_64. Is there any other information I could provide that might be useful?

Comment 3 Josef Bacik 2013-04-29 18:08:09 UTC

Sorry this is just the enospc accounting stuff being noisy, it needs to be reworked.  It shouldn't cause any real problems.  I'll update this bug when I've fixed the problem.

Comment 4 Josh Boyer 2013-07-01 13:55:49 UTC

Josef, did this get fixed upstream?

Comment 5 Josef Bacik 2013-09-23 20:45:32 UTC

Hasn't been fixed upstream yet, is being tracked in an upstream bugzilla tho so closing this bug

https://bugzilla.kernel.org/show_bug.cgi?id=53001