You are not logged in.

#1 2011-11-10 03:22:39

i336
Member
Registered: 2011-01-08
Posts: 11

/ the only supposedly "mounted" partition, fs segfaults / 3.1.0-4

[Update: Rebooting fixed the fs segfaults (were they segfaults?) although the "mounting filesystems... [FAIL]" message is still present.]

I should probably have paid closer attention to the fact that ever since I went from 2.6 to 3.0 Arch has been saying "mounting filesystems [FAIL]" every boot. I tried removing /dev/fd0 and /dev/cdrom but the errors kept continuing. "Oh well," I thought, "something's just interpreting something else's output wrong."

Whoops.

Now, with kernel 3.1.0-4, those error messages have been complemented with some additional bizarreness; observe and ponder:

/root/ + mount
proc on /proc type proc (rw,nosuid,nodev,noexec,relatime)
sys on /sys type sysfs (rw,nosuid,nodev,noexec,relatime)
udev on /dev type devtmpfs (rw,nosuid,relatime,size=10240k,nr_inodes=62747,mode=755)
run on /run type tmpfs (rw,nosuid,nodev,relatime,size=10240k,mode=755)
/dev/sda5 on / type jfs (rw,relatime)
/root/ + cat /etc/mtab
proc /proc proc rw,nosuid,nodev,noexec,relatime
sys /sys sysfs rw,nosuid,nodev,noexec,relatime
udev /dev devtmpfs rw,nosuid,relatime,size=10240k,nr_inodes=62747,mode=755
run /run tmpfs rw,nosuid,nodev,relatime,size=10240k,mode=755
/dev/sda5 / jfs rw,relatime
 /dev/pts devpts rw,relatime,mode=600,ptmxmode=000
 /dev/shm tmpfs rw,relatime
/root/ + cat /proc/mounts # identical to /proc/self/mounts
rootfs / rootfs rw 0 0
proc /proc proc rw,nosuid,nodev,noexec,relatime 0 0
sys /sys sysfs rw,nosuid,nodev,noexec,relatime 0 0
udev /dev devtmpfs rw,nosuid,relatime,size=10240k,nr_inodes=62747,mode=755 0 0
run /run tmpfs rw,nosuid,nodev,relatime,size=10240k,mode=755 0 0
/dev/sda5 / jfs rw,relatime 0 0
none /dev/pts devpts rw,relatime,mode=600,ptmxmode=000 0 0
none /dev/shm tmpfs rw,relatime 0 0
none /dev/shm tmpfs rw,relatime 0 0
/dev/sda1 /boot jfs rw,relatime 0 0
/dev/sda6 /var jfs rw,relatime 0 0
/dev/sda8 /tmp jfs rw,relatime 0 0
/dev/sda9 /home jfs rw,relatime 0 0
/root/ +

Looking around for causes, dmesg contains quite a few potential hairperms except... hi there, I'm on Chrome, I have a bunch of terminals open... I'm online.

A couple of salient portions of http://paste.pocoo.org/show/505609/:

[   23.244682] ------------[ cut here ]------------
[   23.244721] WARNING: at fs/inode.c:884 unlock_new_inode+0x57/0x60()
[   23.244728] Hardware name: 6578PAA
[   23.244732] Modules linked in: lm90 i915 drm_kms_helper drm i2c_algo_bit video loop fuse serio_raw evdev snd_usb_audio pcspkr i2c_i801 button snd_usbmidi_lib iTCO_wdt iTCO_vendor_support ppdev psmouse intel_agp parport_pc parport i2c_core floppy intel_gtt shpchp agpgart processor pci_hotplug snd_intel8x0 snd_ac97_codec snd_rawmidi snd_hwdep snd_seq_dummy snd_seq_oss snd_seq_midi_event snd_seq snd_seq_device snd_pcm_oss snd_pcm snd_timer snd_page_alloc snd_mixer_oss snd soundcore ac97_bus slhc e1000 e100 mii jfs dm_crypt dm_mod usbhid hid uhci_hcd usbcore tulip sd_mod ata_piix ata_generic libata scsi_mod
[   23.244857] Pid: 492, comm: mount Not tainted 3.1.0-4-ARCH #1
[   23.244863] Call Trace:
[   23.244882]  [<c04596ba>] ? printk+0x1d/0x1f
[   23.244943]  [<c01470b2>] warn_slowpath_common+0x72/0xa0
[   23.244954]  [<c0231947>] ? unlock_new_inode+0x57/0x60
[   23.244963]  [<c0231947>] ? unlock_new_inode+0x57/0x60
[   23.244973]  [<c0147102>] warn_slowpath_null+0x22/0x30
[   23.244982]  [<c0231947>] unlock_new_inode+0x57/0x60
[   23.245026]  [<e0b7231b>] ialloc+0x22b/0x270 [jfs]
[   23.245044]  [<e0b5c76a>] jfs_create+0x5a/0x300 [jfs]
[   23.245065]  [<c0225d3a>] ? do_lookup+0x4a/0x340
[   23.245075]  [<c0230452>] ? __d_alloc+0x22/0x120
[   23.245084]  [<c0230508>] ? __d_alloc+0xd8/0x120
[   23.245110]  [<c028cc2a>] ? security_inode_permission+0x1a/0x20
[   23.245120]  [<c0226410>] ? inode_permission+0x40/0xe0
[   23.245132]  [<c022726b>] vfs_create+0x9b/0x110
[   23.245143]  [<c0229253>] do_last+0x743/0x870
[   23.245154]  [<c02294e4>] path_openat+0xa4/0x350
[   23.245164]  [<c01f1561>] ? handle_pte_fault+0xe1/0x8e0
[   23.245175]  [<c02297f1>] do_filp_open+0x31/0x80
[   23.245191]  [<c0234443>] ? alloc_fd+0xa3/0xe0
[   23.245201]  [<c0225ad5>] ? getname_flags+0xf5/0x130
[   23.245224]  [<c021aec6>] do_sys_open+0xe6/0x1b0
[   23.245243]  [<c016fa6f>] ? getnstimeofday+0x4f/0x110
[   23.245254]  [<c021afbe>] sys_open+0x2e/0x40
[   23.245268]  [<c0461f9f>] sysenter_do_call+0x12/0x28
[   23.245276] ---[ end trace e29693b6fc28e267 ]---
[   23.278169] ------------[ cut here ]------------

Some scary statistics:

/proc/self/ + dmesg | grep ', comm: ' | grep mount | wc -l
5
/proc/self/ + dmesg | grep ', comm: ' | grep chrome | wc -l
176
/proc/self/ +

Meep?

Should I run away to kernel26-lts for a little while? This is someone else's PC that I maintain for them every now and again, so relatively, how will I say, "self-contained, un-interactive stability" is kind of particularly important here. tongue

Help greatly appreciated.

-i336

Last edited by i336 (2011-11-10 04:29:30)

Offline

#2 2011-11-10 15:05:44

karol
Archivist
Registered: 2009-05-06
Posts: 25,440

Re: / the only supposedly "mounted" partition, fs segfaults / 3.1.0-4

Can you post your fstab?

Offline

#3 2011-11-10 16:01:02

falconindy
Developer
From: New York, USA
Registered: 2009-10-22
Posts: 4,111
Website

Re: / the only supposedly "mounted" partition, fs segfaults / 3.1.0-4

i336 wrote:

Should I run away to kernel26-lts for a little while? This is someone else's PC that I maintain for them every now and again, so relatively, how will I say, "self-contained, un-interactive stability" is kind of particularly important here. tongue

If stability is important, I suggest using a different FS, such as ext3 or ext4.

Offline

#4 2011-11-10 16:36:31

Leonid.I
Member
From: Aethyr
Registered: 2009-03-22
Posts: 999

Re: / the only supposedly "mounted" partition, fs segfaults / 3.1.0-4

You have jfs, right? I found it VERY unusable in a VM with kernel 3.0 up. Namely, if the qemu process dies due to some reason, I get exactly same errors in dmesg even after successful fsck on boot. Only a second reboot helps. Even btrfs is more reliable IMHO.

Last edited by Leonid.I (2011-11-10 16:36:54)


Arch Linux is more than just GNU/Linux -- it's an adventure
pkill -9 systemd

Offline

Board footer

Powered by FluxBB