On Fri, Jan 22, 2010 at 07:00:29PM +0100, Mattias Säteri wrote: > Hi, > > I ran into a bug when trying to replace a failed device in a RAID-1 > configuration. I did the following: > > 1. Create raid1 fs following the description on the btrfs wiki > (http://btrfs.wiki.kernel.org/index.php/Using_Btrfs_with_Multiple_Devices). > 2. Mount fs and copy some data to it. > 3. Unmount fs and overwrite one of the partitions with junk. > 4. Rescan for btrfs volumes with btrfsctl -a (as expected, finds only > one volume). > 5. Try to add the device resulting in a crash, see below: > > ---------------------------------------------------------------------------- > > # mount -o degraded /dev/sda4 btrfs/ > # btrfs-vol -a /dev/hda1 btrfs/ > Segmentation fault > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740084] ------------[ cut here ]------------ > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740089] last sysfs file: /sys/block/sdc/size > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740087] invalid opcode: 0000 [#1] SMP > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740170] Stack: > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740179] Call Trace: > > Message from sysl...@bamse at Jan 15 22:14:59 ... > kernel:[11517.740228] Code: 44 5e 04 e1 e9 98 fe ff ff be d9 03 00 00 > 48 c7 c7 de 86 35 a0 e8 7e a7 d2 e0 e9 ca fa ff ff 0f 0b eb fe 0f 0b > eb fe 0f 0b eb fe <0f> 0b eb fe 0f 0b eb fe 0f 0b eb fe be 05 04 00 00 > 48 c7 c7 de > > bamse:/mnt# btrfs-vol -a /dev/hda1 btrfs > failed to zero device end -5 > > ------------------------------------------------------------------------------ > > Output from dmesg below: > > [11467.229110] btrfs: allowing degraded mounts > [11517.734662] btrfs allocation failed flags 18, wanted 4096 > [11517.734666] space_info has 8376320 free, is not full > [11517.734667] space_info total=8388608, pinned=0, delalloc=0, may_use=0, > used=1 > 2288, root=0, super=0, reserved=0 > [11517.734670] block group 20971520 has 8388608 bytes, 12288 used 0 pinned 0 > res > erved > [11517.734671] block group has cluster?: no > [11517.734673] 0 blocks of free space at or bigger than bytes is > [11517.740040] btrfs allocation failed flags 20, wanted 4096 > [11517.740042] space_info has 1012105216 free, is not full > [11517.740044] space_info total=1073741824, pinned=0, delalloc=0, may_use=0, > use > d=61636608, root=0, super=0, reserved=0 > [11517.740046] block group 29360128 has 1073741824 bytes, 61636608 used 0 > pinned > 0 reserved > [11517.740048] block group has cluster?: no > [11517.740049] 0 blocks of free space at or bigger than bytes is > [11517.740085] kernel BUG at fs/btrfs/transaction.c:1047! > [11517.740087] invalid opcode: 0000 [#1] SMP > [11517.740089] last sysfs file: /sys/block/sdc/size > [11517.740091] CPU 0 > [11517.740092] Modules linked in: nfnetlink_queue nfnetlink nvidia(P) > xt_multipo > rt nf_conntrack_ipv4 nf_defrag_ipv4 ipt_REJECT xt_tcpudp iptable_filter > ip_table > s xt_iprange xt_state nf_conntrack xt_mark xt_NFQUEUE x_tables > dvbloopback ppdev lp ipv6 nfs reiserfs fuse usb_storage btrfs > zlib_deflate crc32c libcrc32c dvb_usb_dib0700 dib7000p dib7000m > dib0070 dvb_usb mt2266 tuner_xc2028 dib8000 dib3000mc dibx000_common > mxl5007t xc5000 s5h1411 mt2060 lgdt3305 dvb_core nfsd lockd nfs_acl > auth_rpcgss sunrpc exportfs loop snd_hda_codec_realtek snd_hda_intel > snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy > snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event snd_seq > snd_timer snd_seq_device i2c_i801 r8169 thermal psmouse snd parport_pc > parport iTCO_wdt jmicron button mii i2c_core evdev ehci_hcd processor > soundcore serio_raw floppy uhci_hcd thermal_sys snd_page_alloc pcspkr > sr_mod > [11517.740134] Pid: 4582, comm: btrfs-vol Tainted: P > 2.6.32.3 #3 EP43-S3L > [11517.740136] RIP: 0010:[<ffffffffa031b903>] [<ffffffffa031b903>] > btrfs_commit_transaction+0x693/0x6c0 [btrfs] > [11517.740147] RSP: 0018:ffff8801b5063c68 EFLAGS: 00010286 > [11517.740149] RAX: 00000000ffffffe4 RBX: ffff8801b5b3fe10 RCX: > 0000000000000016 > [11517.740151] RDX: 0000000000000000 RSI: ffff8801b5063be8 RDI: > ffff8800af168060 > [11517.740152] RBP: ffff8801b5b3fe10 R08: 0000000000000000 R09: > ffff880235063727 > [11517.740154] R10: 0000000000000001 R11: 0000000000000000 R12: > ffff8800ae211800 > [11517.740155] R13: ffff8801b5b3feb0 R14: 0000000000000001 R15: > 0000000000000001 > [11517.740157] FS: 00007f40569a2730(0000) GS:ffff880028200000(0000) > knlGS:0000000000000000 > [11517.740159] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b > [11517.740161] CR2: 00007f405632e000 CR3: 00000001b5495000 CR4: > 00000000000406f0 > [11517.740162] DR0: 0000000000000000 DR1: 0000000000000000 DR2: > 0000000000000000 > [11517.740164] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: > 0000000000000400 > [11517.740169] Process btrfs-vol (pid: 4582, threadinfo > ffff8801b5062000, task ffff8800bf1e6c90) > [11517.740170] Stack: > [11517.740171] ffff8801c27534e0 ffff8801b5b3fe80 0000000000000000 > 00000000a033dedc > [11517.740173] <0> 0000000000000000 ffff8800bf1e6c90 ffffffff8105c170 > ffff8801b5063ca0 > [11517.740176] <0> ffff8801b5063ca0 0000006200000100 ffff8800aea54800 > ffff88022ba4ab40 > [11517.740179] Call Trace: > [11517.740184] [<ffffffff8105c170>] ? autoremove_wake_function+0x0/0x30 > [11517.740197] [<ffffffffa0342509>] ? btrfs_init_new_device+0x6e9/0xb50 > [btrfs] > [11517.740201] [<ffffffff8108f852>] ? __generic_file_aio_write+0x292/0x490 > [11517.740204] [<ffffffff8108d92b>] ? find_get_page+0x1b/0xb0 > [11517.740206] [<ffffffff8108f1b8>] ? filemap_fault+0xb8/0x400 > [11517.740212] [<ffffffffa0345aea>] ? btrfs_ioctl+0x66a/0xa80 [btrfs] > [11517.740216] [<ffffffff810d0cbf>] ? vfs_ioctl+0x2f/0xb0 > [11517.740218] [<ffffffff810d0e70>] ? do_vfs_ioctl+0x90/0x5d0 > [11517.740221] [<ffffffff810d13f9>] ? sys_ioctl+0x49/0x80 > [11517.740224] [<ffffffff81362ebf>] ? page_fault+0x1f/0x30 > [11517.740227] [<ffffffff8100b52b>] ? system_call_fastpath+0x16/0x1b > [11517.740228] Code: 44 5e 04 e1 e9 98 fe ff ff be d9 03 00 00 48 c7 > c7 de 86 35 a0 e8 7e a7 d2 e0 e9 ca fa ff ff 0f 0b eb fe 0f 0b eb fe > 0f 0b eb fe <0f> 0b eb fe 0f 0b eb fe 0f 0b eb fe be 05 04 00 00 48 c7 > c7 de > [11517.740250] RIP [<ffffffffa031b903>] > btrfs_commit_transaction+0x693/0x6c0 [btrfs] > [11517.740258] RSP <ffff8801b5063c68> > [11517.740260] ---[ end trace 272159a0659829e1 ]--- > > > I use a 2.6.32.3 kernel and btrfs claims to be v0.19-4-gab8fb4c. A > very similar bug was reported on 2.6.30 a while back, so I wonder if > anyone is looking at this already? Is this supposed to work or is it a > known problem? > > BTW I also tried to remove the device from the array with btrfs-vol > -r, but it refuses to go below two devices on raid 1. >
Thanks for the report, I've just posted 3 patches to resolve various issues, including this problem you've posted. Please give them a try and let me know if you have any problems. Thanks, Josef -- To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
