I see this on 3.2.0-38-generic, I have 3x3TB striped volume with LZO, and it suddenly turned dog slow. It is about 90% full atm.
In dmesg I find a bunch of warnings: [2737995.991648] btrfs: block rsv returned -28 [2737995.991649] ------------[ cut here ]------------ [2737995.991655] WARNING: at /build/buildd/linux-3.2.0/fs/btrfs/extent-tree.c:5985 use_block_rsv+0x17c/0x190 [btrfs]() [2737995.991656] Hardware name: Precision WorkStation T7500 [2737995.991657] Modules linked in: ses enclosure usb_storage ufs qnx4 hfsplus hfs minix ntfs vfat msdos fat jfs xfs reiserfs ext2 dm_crypt nfsd nfs lockd fscache auth_rpcgss binfmt_misc nfs_acl sunrpc lp snd_hda_codec_analog snd_hda_intel snd_hda_codec snd_hwdep ppdev snd_pcm dell_wmi parport_pc snd_timer parport i7core_edac sparse_keymap snd shpchp soundcore dcdbas edac_core psmouse serio_raw mac_hid snd_page_alloc btrfs zlib_deflate libcrc32c usbhid hid mxm_wmi radeon mptsas ttm mptscsih firewire_ohci drm_kms_helper mptbase firewire_core tg3 crc_itu_t scsi_transport_sas drm wmi i2c_algo_bit [2737995.991679] Pid: 45809, comm: btrfs-endio-wri Tainted: G W 3.2.0-38-generic #59-Ubuntu [2737995.991680] Call Trace: [2737995.991682] [<ffffffff81067e0f>] warn_slowpath_common+0x7f/0xc0 [2737995.991684] [<ffffffff81067e6a>] warn_slowpath_null+0x1a/0x20 [2737995.991690] [<ffffffffa023162c>] use_block_rsv+0x17c/0x190 [btrfs] [2737995.991697] [<ffffffffa02343fd>] btrfs_alloc_free_block+0x3d/0x200 [btrfs] [2737995.991699] [<ffffffff81122ecb>] ? __set_page_dirty_nobuffers+0x1b/0x20 [2737995.991708] [<ffffffffa0263b8a>] ? read_extent_buffer+0xea/0x180 [btrfs] [2737995.991711] [<ffffffff8165d21e>] ? _raw_spin_lock+0xe/0x20 [2737995.991717] [<ffffffffa0223122>] __btrfs_cow_block+0x122/0x430 [btrfs] [2737995.991722] [<ffffffffa0223526>] btrfs_cow_block+0xf6/0x220 [btrfs] [2737995.991730] [<ffffffffa0272063>] ? btrfs_set_lock_blocking_rw+0x43/0xb0 [btrfs] [2737995.991736] [<ffffffffa022723a>] btrfs_search_slot+0x34a/0x710 [btrfs] [2737995.991743] [<ffffffffa0238988>] btrfs_lookup_file_extent+0x38/0x40 [btrfs] [2737995.991751] [<ffffffffa02534bc>] btrfs_drop_extents+0xec/0x930 [btrfs] [2737995.991753] [<ffffffff8116588c>] ? kmem_cache_alloc+0x10c/0x140 [2737995.991759] [<ffffffffa022251a>] ? btrfs_alloc_path+0x1a/0x20 [btrfs] [2737995.991767] [<ffffffffa0246b63>] insert_reserved_file_extent.constprop.52+0x73/0x270 [btrfs] [2737995.991775] [<ffffffffa0243d4d>] ? start_transaction.part.25+0x7d/0x2a0 [btrfs] [2737995.991783] [<ffffffffa024a72a>] btrfs_finish_ordered_io+0x2ca/0x320 [btrfs] [2737995.991791] [<ffffffffa024a7cd>] btrfs_writepage_end_io_hook+0x4d/0xc0 [btrfs] [2737995.991799] [<ffffffffa025edc9>] end_bio_extent_writepage+0x129/0x180 [btrfs] [2737995.991801] [<ffffffff811adf3d>] bio_endio+0x1d/0x40 [2737995.991809] [<ffffffffa023d2ad>] end_workqueue_fn+0x5d/0xa0 [btrfs] [2737995.991817] [<ffffffffa026c4f0>] worker_loop+0xa0/0x2b0 [btrfs] [2737995.991818] [<ffffffff8165acbc>] ? __schedule+0x3cc/0x6f0 [2737995.991827] [<ffffffffa026c450>] ? check_pending_worker_creates.isra.1+0xf0/0xf0 [btrfs] [2737995.991829] [<ffffffff8108b19c>] kthread+0x8c/0xa0 [2737995.991831] [<ffffffff81667974>] kernel_thread_helper+0x4/0x10 [2737995.991833] [<ffffffff8108b110>] ? flush_kthread_worker+0xa0/0xa0 [2737995.991835] [<ffffffff81667970>] ? gs_change+0x13/0x13 [2737995.991836] ---[ end trace 66ce02ac277ff6d4 ]--- [2738031.380617] use_block_rsv: 1 callbacks suppressed [2738031.380621] btrfs: block rsv returned -28 [2738031.380623] ------------[ cut here ]------------ -- You received this bug notification because you are a member of Ubuntu Bugs, which is subscribed to Ubuntu. https://bugs.launchpad.net/bugs/791992 Title: btrfs in 2.6.38 extremely slow (when near full?) To manage notifications about this bug go to: https://bugs.launchpad.net/ubuntu/+source/linux/+bug/791992/+subscriptions -- ubuntu-bugs mailing list ubuntu-bugs@lists.ubuntu.com https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs
