On Wed, Dec 21, 2011 at 7:35 AM, Christian Brunner <chb@xxxxxx> wrote: > 2011/12/21 Jens Rehpöhler <jens.rehpoehler@xxxxxxxx>: >> Am 19.12.2011 19:51, schrieb Gregory Farnum: >>> On Mon, Dec 19, 2011 at 12:28 AM, Jens Rehpöhler >>> <jens.rehpoehler@xxxxxxxx> wrote: >>>> Good morning !! >>>> >>>> i got the following warning as soon as i use btrfs as underlaying >>>> filesystem (for stability reasons i use ext4). I converted one OSD back >>>> to btrfs to give it another try. >>>> >>>> kernel: 3.2rc6 >>>> ceph: ceph/stable uptodate 0.39-9-g37e7a52-1~bpo60+1 >>>> dist: debian squeeze >>>> >>>> [...] >>>> >>>> Anything to worry about ? >>> This doesn't look familiar to me — if you can reproduce it (or maybe >>> just provide more detail) the btrfs folks will probably be interested. >>> -Greg >> >> Hi Greg, >> >> reproduction is possible, if i destroy one osd und start a fresh resync >> (actual 550 GB of data). >> After a short while i get: >> >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.477676] ------------[ cut here >> ]------------ >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.477727] WARNING: at >> fs/btrfs/inode.c:1969 btrfs_orphan_commit_root+0x7f/0x9b [btrfs]() >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.477772] Hardware name: X8DT3 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.477795] Modules linked in: >> zram(C) dm_mod rbd loop aes_x86_64 aes_generic cbc kvm_intel >> xt_multiport ext2 iptable_filter ip_tables vhost_net macvtap macvlan tun >> brd ebt_arp ebtable_filter ebtables x_tables kvm fuse ceph libceph >> bridge stp llc ext4 jbd2 crc16 tpm_tis tpm joydev pcspkr ioatdma >> tpm_bios processor i2c_i801 thermal_sys serio_raw button i2c_core evdev >> ext3 jbd mbcache btrfs zlib_deflate crc32c libcrc32c sd_mod crc_t10dif >> usbhid hid uhci_hcd ehci_hcd usbcore ixgbe 3w_sas mdio usb_common >> scsi_mod igb dca [last unloaded: scsi_wait_scan] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478130] Pid: 6790, comm: >> btrfs-transacti Tainted: G WC 3.2.0-rc6 #2 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478173] Call Trace: >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478199] [<ffffffff81047257>] >> ? warn_slowpath_common+0x78/0x8c >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478237] [<ffffffffa011ee22>] >> ? btrfs_orphan_commit_root+0x7f/0x9b [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478289] [<ffffffffa01182f3>] >> ? commit_fs_roots+0xa2/0x159 [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478332] [<ffffffff810ece7c>] >> ? kmem_cache_free+0x12/0x9e >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478377] [<ffffffffa0118fcd>] >> ? btrfs_commit_transaction+0x3e2/0x71b [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478423] [<ffffffff81053026>] >> ? del_timer_sync+0x34/0x3e >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478453] [<ffffffff81328237>] >> ? schedule_timeout+0xa8/0xd7 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478482] [<ffffffff81060b69>] >> ? wake_up_bit+0x20/0x20 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478517] [<ffffffffa0113c8d>] >> ? transaction_kthread+0x173/0x22d [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478560] [<ffffffffa0113b1a>] >> ? find_and_setup_root+0xdc/0xdc [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478600] [<ffffffffa0113b1a>] >> ? find_and_setup_root+0xdc/0xdc [btrfs] >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478630] [<ffffffff8106072d>] >> ? kthread+0x7a/0x82 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478658] [<ffffffff813316b4>] >> ? kernel_thread_helper+0x4/0x10 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478687] [<ffffffff810606b3>] >> ? kthread_worker_fn+0x147/0x147 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478716] [<ffffffff813316b0>] >> ? gs_change+0x13/0x13 >> Dec 20 19:45:22 fcmsnode4 kernel: [170010.478742] ---[ end trace >> 3b774941bf3fc467 ]--- >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535021] ------------[ cut here >> ]------------ >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535072] WARNING: at >> fs/btrfs/inode.c:1969 btrfs_orphan_commit_root+0x7f/0x9b [btrfs]() >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535117] Hardware name: X8DT3 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535139] Modules linked in: >> zram(C) dm_mod rbd loop aes_x86_64 aes_generic cbc kvm_intel >> xt_multiport ext2 iptable_filter ip_tables vhost_net macvtap macvlan tun >> brd ebt_arp ebtable_filter ebtables x_tables kvm fuse ceph libceph >> bridge stp llc ext4 jbd2 crc16 tpm_tis tpm joydev pcspkr ioatdma >> tpm_bios processor i2c_i801 thermal_sys serio_raw button i2c_core evdev >> ext3 jbd mbcache btrfs zlib_deflate crc32c libcrc32c sd_mod crc_t10dif >> usbhid hid uhci_hcd ehci_hcd usbcore ixgbe 3w_sas mdio usb_common >> scsi_mod igb dca [last unloaded: scsi_wait_scan] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535471] Pid: 6790, comm: >> btrfs-transacti Tainted: G WC 3.2.0-rc6 #2 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535514] Call Trace: >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535539] [<ffffffff81047257>] >> ? warn_slowpath_common+0x78/0x8c >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535577] [<ffffffffa011ee22>] >> ? btrfs_orphan_commit_root+0x7f/0x9b [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535629] [<ffffffffa01182f3>] >> ? commit_fs_roots+0xa2/0x159 [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535661] [<ffffffff810ece7c>] >> ? kmem_cache_free+0x12/0x9e >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535697] [<ffffffffa0118fcd>] >> ? btrfs_commit_transaction+0x3e2/0x71b [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535743] [<ffffffff81053026>] >> ? del_timer_sync+0x34/0x3e >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535773] [<ffffffff81328237>] >> ? schedule_timeout+0xa8/0xd7 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535803] [<ffffffff81060b69>] >> ? wake_up_bit+0x20/0x20 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535838] [<ffffffffa0113c8d>] >> ? transaction_kthread+0x173/0x22d [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535876] [<ffffffffa0113b1a>] >> ? find_and_setup_root+0xdc/0xdc [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535914] [<ffffffffa0113b1a>] >> ? find_and_setup_root+0xdc/0xdc [btrfs] >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535944] [<ffffffff8106072d>] >> ? kthread+0x7a/0x82 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.535971] [<ffffffff813316b4>] >> ? kernel_thread_helper+0x4/0x10 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.536000] [<ffffffff810606b3>] >> ? kthread_worker_fn+0x147/0x147 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.536028] [<ffffffff813316b0>] >> ? gs_change+0x13/0x13 >> Dec 20 19:48:57 fcmsnode4 kernel: [170225.536054] ---[ end trace >> 3b774941bf3fc468 ]--- >> >> Seems to me, that this happens if the btrfs is under heavy load. > > I was thinking that everyone using ceph on btrfs is hitting this > WARNING sooner or later. > > We have seen this a long time ago: > > http://tracker.newdream.net/issues/563 > > There were several threads on the btrfs list, but there is no solution > yet. Here is one for example: > > http://comments.gmane.org/gmane.comp.file-systems.btrfs/14193 > > The good news is, that I haven't experienced any problems that I could > relate to this warning. (Which doesn't mean that I am happy with the > current state of btrfs). Well the backtrace is different from in that bug, and not one I remember — but I don't know btrfs real well and I'm not sure which functions are the important ones there. It certainly *could* be the same thing; maybe somebody who's written btrfs patches would know... -Greg -- To unsubscribe from this list: send the line "unsubscribe ceph-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
