This has happened to me twice now (symptom: offlineimap and mutt just freeze - all I have on here are some maildirs and the associated offlineimap and mutt state info. It's all under "/shared/brong" which is a snapshot taken from /shared/default just after it was created, owned by my user (brong). /shared is a btrfs filesystem, created with v0.6 (I'm using hg, but I've updated it since and not rebuilt so I lost the exact point at which I built everything, sorry.. I now need to pretty much sync everything and hard reset to unlock this. I can't umount. Anyway, hoping that the following may still be of some use. I know full well it's alpha and I can't expect to trust it, which is why there's nothing on there I can't replace with a few megabytes of re-downloading. Bron. [175658.540000] BUG: unable to handle kernel NULL pointer dereference at virtual address 00000044 [175658.540000] printing eip: [175658.540000] f8cfe55d [175658.540000] *pde = 00000000 [175658.540000] Oops: 0000 [#1] [175658.540000] SMP [175658.540000] Modules linked in: usbhid hid af_packet binfmt_misc rfcomm l2cap capability vboxdrv ppdev ipv6 tun speedstep_centrino cpufreq_userspace cpufreq_stats cpufreq_ondemand freq_table cpufreq_powersave cpufreq_conservative button dock ac sbs battery video container btrfs libcrc32c reiserfs sbp2 lp fuse snd_intel8x0 snd_ac97_codec ac97_bus snd_pcm_oss snd_pcm snd_mixer_oss snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi joydev snd_seq_midi_event snd_seq snd_timer snd_seq_device irda pcmcia crc_ccitt ipw2200 parport_pc parport pcspkr iTCO_wdt iTCO_vendor_support ieee80211 snd soundcore snd_page_alloc psmouse serio_raw ieee80211_crypt yenta_socket hci_usb rsrc_nonstatic bluetooth pcmcia_core shpchp pci_hotplug intel_agp agpgart evdev ext3 jbd mbcache sg sr_mod cdrom sd_mod ata_piix e100 mii ohci1394 ieee1394 ata_generic libata scsi_mod ehci_hcd uhci_hcd usbcore thermal processor fan apparmor commoncap aamatch_pcre [175658.540000] CPU: 0 [175658.540000] EIP: 0060:[<f8cfe55d>] Not tainted VLI [175658.540000] EFLAGS: 00010296 (2.6.22-9-generic #1) [175658.540000] EIP is at btrfs_find_block_group+0x1ed/0x530 [btrfs] [175658.540000] eax: 00000000 ebx: 000c0000 ecx: 00000000 edx: f549eb78 [175658.540000] esi: 00000000 edi: 00000000 ebp: 00000000 esp: f7ff1c34 [175658.540000] ds: 007b es: 007b fs: 00d8 gs: 0000 ss: 0068 [175658.540000] Process btrfs/0 (pid: 4048, ti=f7ff0000 task=f7efea40 task.ti=f7ff0000) [175658.540000] Stack: f2d7f150 e8b1f000 00000000 df915000 f8cfb632 00000067 f7ff1cc4 00001aa1 [175658.540000] 00000000 00000000 dfb88000 f7ff1e34 e0b0a000 dfb8802c dfb88038 00000001 [175658.540000] 00000000 00ff1cbc 00000fff 00000004 e0b77de0 00000006 00000004 00000005 [175658.540000] Call Trace: [175658.540000] [<f8cfb632>] btrfs_search_slot+0xa2/0xc90 [btrfs] [175658.540000] [<f8cfd45c>] btrfs_lookup_block_group+0x5c/0xc0 [btrfs] [175658.540000] [<f8cff36e>] btrfs_alloc_extent+0xace/0x1240 [btrfs] [175658.540000] [<f8cfdb66>] finish_current_insert+0x66/0x100 [btrfs] [175658.540000] [<f8cfdead>] del_pending_extents+0x3d/0x120 [btrfs] [175658.540000] [<f8cffd86>] btrfs_alloc_free_block+0x76/0x140 [btrfs] [175658.540000] [<f8cf966a>] __btrfs_cow_block+0x7a/0x3d0 [btrfs] [175658.540000] [<f8cfca00>] btrfs_realloc_node+0x1e0/0x2e0 [btrfs] [175658.540000] [<f8d0b576>] btrfs_defrag_leaves+0x536/0x6f0 [btrfs] [175658.540000] [<f8d03a25>] btrfs_end_transaction+0x95/0x110 [btrfs] [175658.540000] [<f8d03e6a>] btrfs_start_transaction+0xaa/0x170 [btrfs] [175658.540000] [<f8d048c7>] btrfs_defrag_dirty_roots+0x117/0x180 [btrfs] [175658.540000] [<f8d04a4a>] btrfs_transaction_cleaner+0x11a/0x150 [btrfs] [175658.540000] [<f8d04930>] btrfs_transaction_cleaner+0x0/0x150 [btrfs] [175658.540000] [<c0138251>] run_workqueue+0x81/0x110 [175658.540000] [<c013beb0>] prepare_to_wait+0x20/0x70 [175658.540000] [<c0138c60>] worker_thread+0x0/0x100 [175658.540000] [<c0138d00>] worker_thread+0xa0/0x100 [175658.540000] [<c013bd00>] autoremove_wake_function+0x0/0x50 [175658.540000] [<c0138c60>] worker_thread+0x0/0x100 [175658.540000] [<c013ba42>] kthread+0x42/0x70 [175658.540000] [<c013ba00>] kthread+0x0/0x70 [175658.540000] [<c0105487>] kernel_thread_helper+0x7/0x10 [175658.540000] ======================[175658.540000] Code: 0b 9c 24 ac 00 00 00 8d 48 38 89 54 24 34 89 4c 24 38 0f 84 88 fe ff ff 8b 94 24 ac 00 00 00 8b 8c 24 b0 00 00 00 e8 a3 ee ff ff <3b> 78 44 89 c3 0f 85 6a fe ff ff 8b 40 34 8b 53 38 89 44 24 1c [175658.540000] EIP: [<f8cfe55d>] btrfs_find_block_group+0x1ed/0x530 [btrfs] SS:ESP 0068:f7ff1c34 -- Bron Gondwana brong@fastmail.fm
On Thu, 23 Aug 2007 23:23:56 +1000 "Bron Gondwana" <brong@fastmail.fm> wrote:> This has happened to me twice now (symptom: offlineimap and > mutt just freeze - all I have on here are some maildirs and > the associated offlineimap and mutt state info. > > It's all under "/shared/brong" which is a snapshot taken from > /shared/default just after it was created, owned by my user > (brong). > > /shared is a btrfs filesystem, created with v0.6 (I'm using hg, > but I've updated it since and not rebuilt so I lost the exact > point at which I built everything, sorry.. > > I now need to pretty much sync everything and hard reset to > unlock this. I can't umount. > > Anyway, hoping that the following may still be of some use. I > know full well it's alpha and I can't expect to trust it, which > is why there's nothing on there I can't replace with a few > megabytes of re-downloading.This should be fixed in v0.7 (I hope). There's a related problem where I'm cowing a block in the extent tree at the wrong time, and I'm testing that fix locally. -chris