Dear Junjiro-San, I hesitated a lot to report this bug, since it concerns a pretty old setup but the pain it causes is too strong to live with it.. I know, you're quite short of time for aufs, but hopefully you're able to look into it sometimes.
Due to a motorbike accident, I stayed in a hospital for the last three weeks, and was glad to get back home by the end of last week. Since I still depend on moving by wheelchair for at least a few more weeks, my office a floor down is hard to reach (suffering from a humerus head and tibia head fracture with compartment syndrome). Thanks to an aufs2 based diskless setup, I was able to create a tentative workplace quickly. The setup is a bit dated (openSUSE 11.1), but most of the important stuff is pretty current with the help of openSUSE build service, where I build any needed piece, and the same setup is also used in a company on about 40 desktops with up to 3 heads. The setup is based on kernel 2.6.31, and aufs2-31 as of 20100823 (latest change on that branch). Traditionally, I'm hitting my workstation harder than anybody else around here. Therefor I harvested many crashes quickly, that happen only occasionally on other systems: Jun 22 16:13:08 kuno kernel: [ 9427.321510] ------------[ cut here ]------------ Jun 22 16:13:08 kuno kernel: [ 9427.322459] kernel BUG at /usr/src/packages/BUILD/kernel-desktop-2.6.31.14/linux-2.6.31/mm/truncate.c:446! Jun 22 16:13:08 kuno kernel: [ 9427.322459] invalid opcode: 0000 [#1] PREEMPT SMP Jun 22 16:13:08 kuno kernel: [ 9427.322459] last sysfs file: /sys/devices/system/cpu/cpu1/cache/index2/shared_cpu_map Jun 22 16:13:08 kuno kernel: [ 9427.322459] Modules linked in: snd_usb_audio snd_usb_lib snd_rawmidi appletalk ax25 ipx p8022 psnap llc p8023 nls_utf8 autofs4 snd_pcm_oss binfmt_misc snd_mixer_oss snd_seq snd_seq_device vmnet vmblock vmci vmmon cpufreq_conservative cpufreq_userspace cpufreq_powersave powernow_k8 fuse loop nvidia(P) snd_hda_codec_realtek wmi thermal button processor thermal_sys snd_hda_intel snd_hda_codec snd_hwdep snd_pcm snd_timer shpchp snd pcspkr pci_hotplug snd_page_alloc i2c_nforce2 aufs squashfs nfs lockd fscache nfs_acl auth_rpcgss sunrpc dm_mod brd af_packet ide_pci_generic amd74xx ide_core ata_generic pata_amd forcedeth parport_pc parport Jun 22 16:13:08 kuno kernel: [ 9427.322459] Jun 22 16:13:08 kuno kernel: [ 9427.322459] Pid: 13524, comm: kmozillahelper Tainted: P (2.6.31.14-6-desktop # 1) To Be Filled By O.E.M. Jun 22 16:13:08 kuno kernel: [ 9427.322459] EIP: 0060:[<c02e391d>] EFLAGS: 00210202 CPU: 0 Jun 22 16:13:08 kuno kernel: [ 9427.322459] EIP is at invalidate_inode_pages2_range+0x1dd/0x270 Jun 22 16:13:08 kuno kernel: [ 9427.322459] EAX: 00000001 EBX: 00000000 ECX: 00000000 EDX: 00000000 Jun 22 16:13:08 kuno kernel: [ 9427.322459] ESI: 00000001 EDI: f5167bc0 EBP: f0109de4 ESP: f0109d70 Jun 22 16:13:08 kuno kernel: [ 9427.322459] DS: 007b ES: 007b FS: 00d8 GS: 00e0 SS: 0068 Jun 22 16:13:08 kuno kernel: [ 9427.322459] Process kmozillahelper (pid: 13524, ti=f0108000 task=f01aaf00 task.ti=f0108000) Jun 22 16:13:08 kuno kernel: [ 9427.322459] Stack: Jun 22 16:13:08 kuno kernel: [ 9427.322459] 00000000 00000000 00000000 ffffffff e6d2471c 00000001 00000000 00000000 Jun 22 16:13:08 kuno kernel: [ 9427.322459] <0> 00000000 0000000e 00000000 f5167bc0 f5167be0 f5167c00 f5167c20 f5167c40 Jun 22 16:13:08 kuno kernel: [ 9427.322459] <0> f5167c60 f5167c80 f5167ca0 f5167cc0 f5167ce0 f5167d00 f5167d20 f5167d40 Jun 22 16:13:08 kuno kernel: [ 9427.322459] Call Trace: Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<c02e39cc>] invalidate_inode_pages2+0x1c/0x40 Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<f800cbb4>] nfs_invalidate_mapping_nolock+0x24/0x110 [nfs] Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<f800e254>] nfs_revalidate_mapping+0xd4/0xe0 [nfs] Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<f800a55b>] nfs_file_mmap+0x4b/0x90 [nfs] Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<f80c36a2>] aufs_mmap+0x172/0x290 [aufs] Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<c02fd5d5>] mmap_region+0x375/0x4d0 Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<c02fd9a5>] do_mmap_pgoff+0x275/0x370 Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<c020934c>] sys_mmap2+0xdc/0x100 Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<c02042b3>] sysenter_do_call+0x12/0x28 Jun 22 16:13:08 kuno kernel: [ 9427.322459] [<ffffe424>] 0xffffe424 Jun 22 16:13:08 kuno kernel: [ 9427.322459] Code: 0c c1 e2 0c c7 44 24 08 00 00 00 00 c7 04 24 00 10 00 00 c7 44 24 04 0 0 00 00 00 e8 ce 20 01 00 8b 47 08 85 c0 0f 88 72 ff ff ff <0f> 0b eb fe 8d b4 26 00 00 00 00 ba 0e 00 00 00 89 f8 e8 1c 37 Jun 22 16:13:08 kuno kernel: [ 9427.322459] EIP: [<c02e391d>] invalidate_inode_pages2_range+0x1dd/0x270 SS:ESP 0068:f0109d70 Jun 22 16:13:08 kuno kernel: [ 9427.564924] ---[ end trace 51dab74ed98a9231 ]--- The offending processes change, but the crash is always the same as above: kio_thumbnail, nscd, kmozillahelper, kwalletd. Before collecting data for this report, I updated aufs2 to the current git version. With default config (debug enabled), the system didn't boot successfully, due to a triggering DyDbgSize in dy_vmop() in dynop.c:226 (I have photos of this issue only). Disabling aufs2 debug "resolved" this for me, but doesn't raise confidence.. The kernel and aufs2 module are built here: https://build.opensuse.org/project/show?project=home%3Afrispete%3Akernel-2.6.31 Again, I know, this concerns quite outdated sstuff, but it is still in heavy use, and a cure for this issue is very appreciated. Junjiro-San, if you need any more details, let me know. Thanks in advance, Pete ------------------------------------------------------------------------------ Simplify data backup and recovery for your virtual environment with vRanger. Installation's a snap, and flexible recovery options mean your data is safe, secure and there when you need it. Data protection magic? Nope - It's vRanger. Get your free trial download today. http://p.sf.net/sfu/quest-sfdev2dev
