FS#24598 - [initscripts] during boot if mounting a partition fails, we should try to keep mounting others

Attached to Project: Arch Linux
Opened by mark (mmm) - Monday, 06 June 2011, 10:37 GMT
Last edited by Tom Gundersen (tomegun) - Sunday, 19 June 2011, 17:01 GMT
Task Type Bug Report
Category Packages: Core
Status Closed
Assigned To Tom Gundersen (tomegun)
Architecture All
Severity High
Priority Normal
Reported Version
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 1
Private No

Details

Description:

with 2.6.39 kernel and initscripts from testing, I can get in conditions when mounting btrfs partition segfaults (I already reported upstream - http://article.gmane.org/gmane.comp.file-systems.btrfs/11135 ).

After this, the boot process continues, the issue is: after the crash, other partitions from fstab don't get mounted. Though there's no error on them and I can flawlessly mount them manually.

What I think happens: the mount process stops (crashes?)
What should happen: mounting the corrupted partition should fail (/home in my case), but other partitions should be probed normally (/var /tmp ...)

Steps to reproduce:
during boot, ensure mounting one partition crashes :P, then other parts are not mounted.

Thanks, Mark
This task depends upon

Closed by  Tom Gundersen (tomegun)
Sunday, 19 June 2011, 17:01 GMT
Reason for closing:  Upstream
Comment by mark (mmm) - Monday, 06 June 2011, 10:38 GMT
sorry, forgot to attach dmesg:

[ 56.994241] loop: module loaded
[ 57.172283] Btrfs loaded
[ 57.191655] device label store devid 1 transid 26106 /dev/dm-3
[ 57.218783] device label home devid 1 transid 450932 /dev/dm-2
[ 57.459448] scsi 4:0:0:0: Direct-Access Generic- Multi-Card 1.00 PQ: 0 ANSI: 0 CCS
[ 57.460293] sd 4:0:0:0: Attached scsi generic sg1 type 0
[ 57.467030] sd 4:0:0:0: [sdb] Attached SCSI removable disk
[ 61.585618] EXT4-fs (sda4): warning: checktime reached, running e2fsck is recommended
[ 61.671534] EXT4-fs (sda4): re-mounted. Opts: (null)
[ 62.211037] device label home devid 1 transid 450932 /dev/mapper/homeDevice
[ 62.212058] btrfs: force lzo compression
[ 65.335194] ------------[ cut here ]------------
[ 65.335308] kernel BUG at fs/btrfs/inode.c:4676!
[ 65.335406] invalid opcode: 0000 [#1] PREEMPT SMP
[ 65.335532] last sysfs file: /sys/devices/virtual/bdi/btrfs-1/uevent
[ 65.337833] Modules linked in: btrfs zlib_deflate crc32c libcrc32c loop uas ums_realtek uvcvideo usb_storage msr videodev media btusb bluetooth sbs sbshc arc4 ecb b43 mac80211 joydev cfg80211 ssb mmc_core pcmcia sg fuse tg3 uhci_hcd ideapad_laptop evdev sparse_keymap psmouse pcspkr snd_hda_codec_realtek iTCO_wdt iTCO_vendor_support rfkill serio_raw ehci_hcd snd_hda_intel pcmcia_core i2c_i801 libphy usbcore ac wmi battery thermal snd_hda_codec snd_hwdep snd_pcm snd_timer snd soundcore snd_page_alloc acpi_cpufreq freq_table processor mperf sha256_generic sha512_generic ext4 mbcache jbd2 crc16 cryptd aes_i586 aes_generic xts gf128mul dm_crypt dm_mod sd_mod ata_piix libata scsi_mod i915 drm_kms_helper drm i2c_algo_bit button i2c_core video intel_agp intel_gtt agpgart
[ 65.337833]
[ 65.337833] Pid: 883, comm: mount Not tainted 2.6.39-ARCH #1 LENOVO 41875QG /Kuril
[ 65.337833] EIP: 0060:[<f9604072>] EFLAGS: 00010282 CPU: 1
[ 65.337833] EIP is at btrfs_add_link+0x172/0x200 [btrfs]
[ 65.337833] EAX: ffffffef EBX: ef448908 ECX: 00000119 EDX: 00000111
[ 65.337833] ESI: 004255d9 EDI: 00000020 EBP: eec77ba4 ESP: eec77b48
[ 65.337833] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068
[ 65.337833] Process mount (pid: 883, ti=eec76000 task=f4c8c450 task.ti=eec76000)
[ 65.337833] Stack:
[ 65.396712] 00000020 004255d9 00000000 eec77b87 00000001 0000e55e 00000000 f960d6f8
[ 65.396712] eec77b88 eec77b8c eec77b90 eec77b94 ef472000 f5076800 ef448ba8 6f43c090
[ 65.396712] 000046ab 01000000 00000000 00000000 00000000 0046ab6f 00000000 eec77c0c
[ 65.396712] Call Trace:
[ 65.396712] [<f960d6f8>] ? btrfs_inode_ref_index+0xd8/0xe0 [btrfs]
[ 65.396712] [<f962cfcf>] add_inode_ref+0x28f/0x320 [btrfs]
[ 65.396712] [<f962de69>] replay_one_buffer+0x239/0x320 [btrfs]
[ 65.396712] [<f961cc97>] ? alloc_extent_buffer+0x77/0x3a0 [btrfs]
[ 65.396712] [<f962b7a9>] walk_down_log_tree+0x1d9/0x370 [btrfs]
[ 65.396712] [<f962b9d9>] walk_log_tree+0x99/0x1c0 [btrfs]
[ 65.396712] [<f962f2fa>] btrfs_recover_log_trees+0x1da/0x2a0 [btrfs]
[ 65.396712] [<f962dc30>] ? replay_one_dir_item+0xb0/0xb0 [btrfs]
[ 65.396712] [<f95f6749>] open_ctree+0x1129/0x1490 [btrfs]
[ 65.396712] [<c11ac7a9>] ? strlcpy+0x39/0x50
[ 65.396712] [<f95d756b>] btrfs_mount+0x4ab/0x5b0 [btrfs]
[ 65.396712] [<c1109d31>] mount_fs+0x31/0x170
[ 65.396712] [<c11207ac>] vfs_kern_mount+0x4c/0x90
[ 65.396712] [<c1120b49>] do_kern_mount+0x39/0xd0
[ 65.396712] [<c1121e31>] do_mount+0x161/0x700
[ 65.396712] [<c11226f6>] sys_mount+0x66/0xa0
[ 65.396712] [<c1330edf>] sysenter_do_call+0x12/0x28
[ 65.396712] Code: 44 24 08 00 00 00 00 89 4c 24 0c 8b 4d 08 89 34 24 e8 73 cc fe ff 85 c0 0f 84 f0 fe ff ff 8b 5d f4 8b 75 f8 8b 7d fc 89 ec 5d c3 <0f> 0b 8b 81 d8 fe ff ff 8d 55 e3 b9 11 00 00 00 89 d7 05 03 01
[ 65.396712] EIP: [<f9604072>] btrfs_add_link+0x172/0x200 [btrfs] SS:ESP 0068:eec77b48
[ 65.397464] ---[ end trace 5f278c10a67bc917 ]---
[ 65.519660] Adding 2561304k swap on /dev/mapper/swapDevice. Priority:-1 extents:1 across:2561304k
[ 67.243199] microcode: CPU0 sig=0x106c2, pf=0x4, revision=0x20a
[ 67.292031] microcode: CPU1 sig=0x106c2, pf=0x4, revision=0x20a
[ 67.298402] microcode: Microcode Update Driver: v2.00 <tigran@aivazian.fsnet.co.uk>, Peter Oruba
[ 67.305857] microcode: CPU0 updated to revision 0x218, date = 2009-04-10
[ 67.315268] microcode: CPU1 updated to revision 0x218, date = 2009-04-10
[ 70.985116] Bluetooth: BNEP (Ethernet Emulation) ver 1.3
[ 71.152025] Bluetooth: RFCOMM TTY layer initialized
[ 71.156497] Bluetooth: RFCOMM socket layer initialized
[ 71.160691] Bluetooth: RFCOMM ver 1.11
[ 118.324105] ip_tables: (C) 2000-2006 Netfilter Core Team
[ 118.574729] nf_conntrack version 0.5.0 (16384 buckets, 65536 max)
Comment by Tom Gundersen (tomegun) - Monday, 06 June 2011, 23:07 GMT
I'll look into this, but my first reaction is that this should be reported upstream to util-linux, as we simply do "mount -a".
Comment by mark (mmm) - Tuesday, 07 June 2011, 09:52 GMT
Thank you Tom. I think you are right, reported upstream: http://article.gmane.org/gmane.linux.utilities.util-linux-ng/4248

Loading...