On Thu, Oct 23, 2014 at 05:28:58PM -0700, Robert White wrote: > Is this related to your 5k snapshot drive and your attempt to go > back kernel revs from 3.17.0 etc? This filesystem has four subvolumes: a mostly empty root subvolume, one containing ~13TB of data, and two read-write snapshot subvolumes taken from the big data subvolume. I have a dozen or so btrfs filesystems representing a variety of workloads. One of them blows up about once a week, usually due to some bug that was fixed a few days before. :-/ > I see that you are using 3.17.1 kernel. Are you also up to the 3.17 > version of the btrfs tools? I was running tools 3.16.2, but I'll build 3.17 now that I found the git repo it lives in. :-P > You may be in deep error land from the long use of 3.10... that > said, the --init-csum-tree or --init-extent-tree options may be your > friend here. The backtrace shows you are in "open_ctree" so the > former is more likely the better bet. This filesystem was built on 3.12 or 3.14. I build stable kernels the same day they come out, so this machine is reasonably up to date. Now that I think about it, my 3.17.1-zb64 kernel also has these commits in it: d379730 Revert "Btrfs: race free update of commit root for ro snapshots" 4238302 Btrfs: fix race in WAIT_SYNC ioctl 75bfb9a Btrfs: cleanup error handling in build_backref_tree bbe9051 Btrfs: fix build_backref_tree issue with multiple shared blocks 32be3a1 btrfs: Fix the wrong condition judgment about subset extent map 1d52c78 Btrfs: try not to ENOSPC on log replay f6acfd5 Btrfs: don't do async reclaim during log replay e6c4efd btrfs: Fix and enhance merge_extent_mapping() to insert best fitted extent map 4d1a40c Btrfs: fix up bounds checking in lseek 78a017a Btrfs: add missing compression property remove in btrfs_ioctl_setflags 12b894c btrfs: Fix a deadlock in btrfs_dev_replace_finishing() 0b4699d btrfs: don't go readonly on existing qgroup items 2fad4e8 btrfs: wake up transaction thread from SYNC_FS ioctl These came from a list posted by Chris recently for the stable kernels. > Do make _sure_ you are using a fairly recent (3.14.x at least?) > version of btrfs tools. You might want to download and compile the > latest (3.17) of the tools for this task even if you don't feel > comfortable installing them (without an rpm etc). > > On 10/23/2014 04:16 PM, Zygo Blaxell wrote: > >I attempted to run btrfs check --repair, but it got stuck spinning > >in what appeared to be an infinite loop. strace and ltrace revealed > >nothing, and gdb wasn't particularly helpful, so I rebuilt btrfs with > >debug symbols and tried again. > > > >Now I get this from btrfs check: > > > > Couldn't map the block 17716610236416 > > No mapping for 17716610236416-17716610252800 > > Couldn't map the block 17716610236416 > > Check tree block failed, want=17716610236416, have=0 > > read block failed check_tree_block > > Couldn't read chunk root > > > >Mount fails too: > > > > Oct 23 18:19:38 testhost kernel: [ 388.193783] BTRFS: device label vgs2-md0 devid 3 transid 282186 /dev/dm-11 > > Oct 23 18:19:38 testhost kernel: [ 388.232892] BTRFS: device label vgs2-md0 devid 1 transid 282186 /dev/mapper/md15 > > Oct 23 18:19:38 testhost kernel: [ 388.233305] BTRFS: device label vgs2-md0 devid 2 transid 282186 /dev/mapper/md16 > > Oct 23 18:19:38 testhost kernel: [ 388.234459] BTRFS: device label vgs2-md0 devid 4 transid 282186 /dev/mapper/md18 > > Oct 23 18:19:38 testhost kernel: [ 388.759456] BTRFS info (device dm-12): disk space caching is enabled > > Oct 23 18:19:38 testhost kernel: [ 388.759462] BTRFS: has skinny extents > > Oct 23 18:19:38 testhost kernel: [ 388.760576] BTRFS critical (device dm-12): unable to find logical 17716610236416 len 4096 > > Oct 23 18:19:38 testhost kernel: [ 388.760733] ------------[ cut here ]------------ > > Oct 23 18:19:38 testhost kernel: [ 388.760807] kernel BUG at fs/btrfs/inode.c:1659! > > Oct 23 18:19:38 testhost kernel: [ 388.760880] invalid opcode: 0000 [#1] PREEMPT SMP > > Oct 23 18:19:38 testhost kernel: [ 388.761063] Modules linked in: tun cpufreq_userspace cpufreq_stats cpufreq_powersave cpufreq_conservative softdog nfsd auth_rpcgss nfs_acl nfs lockd fscache sunrpc dummy ipt_MASQUERADE xt_nat xt_tcpudp xt_state iptable_mangle nf_log_ipv4 nf_log_common xt_LOG xt_limit iptable_nat nf_conntrack_ipv4 nf_defrag_ipv4 nf_nat_ipv4 nf_nat nf_conntrack ip6table_filter ip6_tables iptable_filter ip_tables x_tables sch_fq_codel tcp_illinois dm_crypt snd_hda_codec_realtek snd_hda_codec_generic snd_hda_intel snd_hda_controller snd_hda_codec snd_hwdep snd_pcm snd_seq snd_seq_device snd_timer kvm_amd eeepc_wmi snd kvm asus_wmi sparse_keymap rfkill soundcore evdev pcspkr i2c_piix4 parport_pc i2c_core acpi_cpufreq k10temp parport rtc_cmos video processor wmi button thermal_sys k8temp hwmon_vid hwmon btrfs xor raid6_pq dm_mod raid1 md_mod af_packet ipv6 nbd sg uas crct10dif_pclmul crc32_pclmul crc32c_intel ghash_clmulni_intel aesni_intel aes_x86_64 lrw gf128mul glu > e > _helper ablk_helper cryptd microc > > Oct 23 18:19:38 testhost kernel: ode r8169 mii firmware_class ehci_pci ohci_pci ohci_hcd ehci_hcd > > Oct 23 18:19:38 testhost kernel: [ 388.765409] CPU: 0 PID: 25132 Comm: mount Tainted: G W 3.17.1-zb64+ #1 > > Oct 23 18:19:38 testhost kernel: [ 388.765516] Hardware name: System manufacturer System Product Name/A55BM-E, BIOS 0902 11/14/2013 > > Oct 23 18:19:38 testhost kernel: [ 388.765625] task: ffff8800a3108000 ti: ffff8804083c8000 task.ti: ffff8804083c8000 > > Oct 23 18:19:38 testhost kernel: [ 388.765733] RIP: 0010:[<ffffffffc027ea80>] [<ffffffffc027ea80>] btrfs_merge_bio_hook+0x80/0x90 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.765905] RSP: 0018:ffff8804083cb8b8 EFLAGS: 00010282 > > Oct 23 18:19:38 testhost kernel: [ 388.765979] RAX: 00000000ffffffea RBX: 0000000000001000 RCX: 0000000000000000 > > Oct 23 18:19:38 testhost kernel: [ 388.766055] RDX: 0000000000000001 RSI: ffffffff8179e4f9 RDI: ffffffff810ca45a > > Oct 23 18:19:38 testhost kernel: [ 388.766135] RBP: ffff8804083cb8d8 R08: 0000000000000000 R09: ffff8800000bc1a0 > > Oct 23 18:19:38 testhost kernel: [ 388.766211] R10: ffff8800000b9cc0 R11: 000000000000b7c0 R12: 0000000000001000 > > Oct 23 18:19:38 testhost kernel: [ 388.766287] R13: ffff8803f6ca30e8 R14: 000000080e7c2148 R15: ffff8803fae7cbf8 > > Oct 23 18:19:38 testhost kernel: [ 388.766363] FS: 00007fdb1e9bd800(0000) GS:ffff88041ec00000(0000) knlGS:0000000000000000 > > Oct 23 18:19:38 testhost kernel: [ 388.766470] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b > > Oct 23 18:19:38 testhost kernel: [ 388.766544] CR2: 00007fff6ae70ec8 CR3: 00000003feb6c000 CR4: 00000000000407f0 > > Oct 23 18:19:38 testhost kernel: [ 388.766620] Stack: > > Oct 23 18:19:38 testhost kernel: [ 388.766690] ffff8804083cb8d8 0000000000001000 ffff8804083cbb28 0000000000001000 > > Oct 23 18:19:38 testhost kernel: [ 388.766942] ffff8804083cb938 ffffffffc0299539 ffff8803fae7cbf8 0000002000000000 > > Oct 23 18:19:38 testhost kernel: [ 388.767193] 0000000000000000 ffffea000df7c2d0 ffff880406cb0330 0000101cf8429000 > > Oct 23 18:19:38 testhost kernel: [ 388.767444] Call Trace: > > Oct 23 18:19:38 testhost kernel: [ 388.767541] [<ffffffffc0299539>] submit_extent_page.isra.34+0x159/0x1f0 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.767672] [<ffffffffc029af60>] __do_readpage+0x470/0x770 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.767770] [<ffffffffc0299ed0>] ? repair_io_failure+0x200/0x200 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.767864] [<ffffffffc0271310>] ? verify_parent_transid+0x210/0x210 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.767963] [<ffffffffc0295602>] ? btrfs_lookup_ordered_extent+0x82/0xd0 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768093] [<ffffffffc029b320>] __extent_read_full_page+0xc0/0xd0 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768188] [<ffffffffc0271310>] ? verify_parent_transid+0x210/0x210 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768282] [<ffffffffc0271310>] ? verify_parent_transid+0x210/0x210 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768381] [<ffffffffc029d9d3>] read_extent_buffer_pages+0x253/0x330 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768506] [<ffffffffc0271310>] ? verify_parent_transid+0x210/0x210 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768601] [<ffffffffc02730c1>] btree_read_extent_buffer_pages.constprop.120+0xb1/0x110 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768728] [<ffffffffc02737aa>] read_tree_block+0x3a/0x60 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768822] [<ffffffffc0277bbd>] open_ctree+0x12cd/0x1f00 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.768904] [<ffffffff813c724a>] ? disk_name+0xba/0xc0 > > Oct 23 18:19:38 testhost kernel: [ 388.768993] [<ffffffffc024d403>] btrfs_mount+0x6d3/0x9a0 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.769077] [<ffffffff811c8ec3>] ? alloc_pages_current+0xb3/0x180 > > Oct 23 18:19:38 testhost kernel: [ 388.769161] [<ffffffff811f6443>] mount_fs+0x43/0x1b0 > > Oct 23 18:19:38 testhost kernel: [ 388.769240] [<ffffffff81211e24>] vfs_kern_mount+0x74/0x130 > > Oct 23 18:19:38 testhost kernel: [ 388.769319] [<ffffffff81214292>] do_mount+0x262/0xb40 > > Oct 23 18:19:38 testhost kernel: [ 388.769397] [<ffffffff8117e56e>] ? __get_free_pages+0xe/0x50 > > Oct 23 18:19:38 testhost kernel: [ 388.769473] [<ffffffff81213eba>] ? copy_mount_options+0x3a/0x160 > > Oct 23 18:19:38 testhost kernel: [ 388.769550] [<ffffffff81214e4e>] SyS_mount+0x8e/0xe0 > > Oct 23 18:19:38 testhost kernel: [ 388.769627] [<ffffffff817a842d>] system_call_fastpath+0x1a/0x1f > > Oct 23 18:19:38 testhost kernel: [ 388.769702] Code: c9 45 31 c0 89 fe 48 89 c7 4c 89 65 e8 e8 99 79 02 00 85 c0 78 15 4c 01 e3 31 c0 48 3b 5d e8 0f 97 c0 48 83 c4 10 5b 41 5c 5d c3 <0f> 0b 66 66 66 66 66 2e 0f 1f 84 00 00 00 00 00 66 66 66 66 90 > > Oct 23 18:19:38 testhost kernel: [ 388.772243] RIP [<ffffffffc027ea80>] btrfs_merge_bio_hook+0x80/0x90 [btrfs] > > Oct 23 18:19:38 testhost kernel: [ 388.772373] RSP <ffff8804083cb8b8> > > Oct 23 18:19:38 testhost kernel: [ 388.772490] ---[ end trace 40d6c9d5d219b0fe ]--- > > > >Before I mkfs and restore, I'd like to try repairing it. Any suggestions? > > > > -- > To unsubscribe from this list: send the line "unsubscribe linux-btrfs" in > the body of a message to majordomo@xxxxxxxxxxxxxxx > More majordomo info at http://vger.kernel.org/majordomo-info.html >
Attachment:
signature.asc
Description: Digital signature
