General protection fault

[Boldie]

I have received a general protection fault. Maybe this is of interest, for me it is a problem, because I can not rsync anymore. It crashes ever time I am doing it. I have installed the version from Ubuntu reposiroy, version 0.6.3-3~trusty.

[ 828.259890] general protection fault: 0000 [#1] SMP
[ 828.259913] Modules linked in: x86_pkg_temp_thermal intel_powerclamp coretemp kvm_intel kvm crct10dif_pclmul ast crc32_pclmul ghash_clmulni_intel cryptd ttm serio_raw drm_kms_helper drm syscopyarea sysfillrect sysimgblt joydev shpchp ipmi_si mac_hid video lpc_ich lp parport zfs(POX) zunicode(POX) zavl(POX) zcommon(POX) znvpair(POX) spl(OX) usb_storage hid_generic usbhid hid ahci igb i2c_algo_bit dca psmouse ptp libahci pps_core
[ 828.260063] CPU: 1 PID: 2056 Comm: rsync Tainted: P OX 3.13.0-40-generic #69-Ubuntu
[ 828.260089] Hardware name: ASUSTeK COMPUTER INC. P9D-I Series/P9D-I Series, BIOS 0702 03/28/2014
[ 828.260114] task: ffff88021ecde000 ti: ffff880212dc4000 task.ti: ffff880212dc4000
[ 828.260136] RIP: 0010:[] [] spl_kmem_cache_alloc+0x3b/0xd90 [spl]
[ 828.260170] RSP: 0018:ffff880212dc5698 EFLAGS: 00010246
[ 828.260185] RAX: 0000000100000000 RBX: 0000000000000230 RCX: ffffffffa02584e0
[ 828.260206] RDX: ffff88015089ce58 RSI: 0000000000000230 RDI: f6894156415741e5
[ 828.260226] RBP: ffff880212dc5748 R08: 2ae822b9004922b0 R09: 2ae822b9004922b0
[ 828.260246] R10: ffff88022242d8b0 R11: ffffffffa00d431b R12: 0000000000000001
[ 828.260267] R13: f689415641574a4d R14: ffff8800412c21e0 R15: f6894156415741e5
[ 828.260287] FS: 00007f97b7093740(0000) GS:ffff88022fd00000(0000) knlGS:0000000000000000
[ 828.260310] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 828.260327] CR2: 00007f97b7097000 CR3: 00000002210a6000 CR4: 00000000000407e0
[ 828.260347] Stack:
[ 828.260354] ffffffffa01d93a6 0000000025032aa0 000000979f2bc000 ffffc9006299f000
[ 828.260379] ffffc90025032aa0 0000000000001000 000000100000001b ffff880226c25d40
[ 828.260403] ffff880124c03020 000000979f2bd000 ffff880124c03000 ffffc900629a0000
[ 828.260427] Call Trace:
[ 828.260465] [] ? __vdev_disk_physio+0x376/0x3d0 [zfs]
[ 828.260506] [] zio_buf_alloc+0x23/0x30 [zfs]
[ 828.260531] [] arc_get_data_buf.isra.19+0x345/0x4b0 [zfs]
[ 828.260558] [] arc_buf_alloc+0xdc/0x110 [zfs]
[ 828.260583] [] arc_read+0x36a/0x8f0 [zfs]
[ 828.260602] [] ? mutex_lock+0x12/0x2f
[ 828.260627] [] ? dbuf_rele_and_unlock+0x250/0x250 [zfs]
[ 828.260655] [] dbuf_read+0x23a/0x890 [zfs]
[ 828.260681] [] dmu_spill_hold_by_dnode+0x44/0x160 [zfs]
[ 828.260710] [] dmu_spill_hold_existing+0x14b/0x160 [zfs]
[ 828.260745] [] sa_get_spill.part.8+0x1e/0x70 [zfs]
[ 828.260777] [] sa_attr_op+0x3f4/0x440 [zfs]
[ 828.260807] [] sa_lookup+0x42/0x60 [zfs]
[ 828.260829] [] ? __cv_init+0x5b/0x110 [spl]
[ 828.260864] [] zfs_dirent_lock+0x3b5/0x570 [zfs]
[ 828.260900] [] zfs_get_xattrdir+0x120/0x1a0 [zfs]
[ 828.260924] [] ? nvlist_common.part.102+0xe2/0x1e0 [znvpair]
[ 828.260947] [] ? mutex_lock+0x12/0x2f
[ 828.260982] [] zfs_lookup+0x1bd/0x320 [zfs]
[ 828.261019] [] __zpl_xattr_get+0x76/0x210 [zfs]
[ 828.261054] [] zpl_xattr_get+0x5c/0x120 [zfs]
[ 828.261090] [] zpl_get_acl+0xc6/0x240 [zfs]
[ 828.261803] [] zpl_xattr_acl_get+0x41/0xf0 [zfs]
[ 828.262515] [] zpl_xattr_acl_get_default+0x12/0x20 [zfs]
[ 828.263215] [] generic_getxattr+0x4c/0x70
[ 828.263905] [] vfs_getxattr+0x88/0xb0
[ 828.264569] [] getxattr+0xab/0x1d0
[ 828.265213] [] ? final_putname+0x22/0x50
[ 828.265838] [] ? putname+0x29/0x40
[ 828.266438] [] ? user_path_at_empty+0x5f/0x90
[ 828.267032] [] ? lockref_put_or_lock+0x50/0x80
[ 828.267614] [] ? dput+0x4c/0x180
[ 828.268183] [] SyS_getxattr+0x55/0x90
[ 828.268738] [] system_call_fastpath+0x1a/0x1f
[ 828.269284] Code: 41 55 41 54 53 89 f3 48 81 ec 88 00 00 00 f6 05 0b 47 01 00 01 74 0d f6 05 0b 47 01 00 08 0f 85 54 01 00 00 4d 8d af 68 08 00 00 41 ff 87 68 08 00 00 41 f6 87 48 08 00 00 80 0f 84 8f 00 00
[ 828.270476] RIP [] spl_kmem_cache_alloc+0x3b/0xd90 [spl]
[ 828.271065] RSP
[ 828.297974] ---[ end trace 18af8c3532c80348 ]---

[dweeezil]

Assuming you have xattr=as you've got a corrupted dnode. Should be fixed in current master and the next release.

[Boldie]

Yes, I am using xattr=as, so this can be possible. Is there a way to find out, which dnode is affected and can I fix it?

[dweeezil]

@Boldie You should be able to get the information you need from #2932 or any of the other related issues, a few of which are #2663, #2700, #2701, #2717, #2863 and #2884.

[Boldie]

Ok, tanks for the information. I have the luxury situations to have a backup of the data, so I decided to set it up from sratch and set xattr=on which should work as I have read. I am looking forward for the next version and hopefully I can switch it back to sa then. Maybe there is a reliable way to convert the xattr=on to xattr=sa thenn.

[behlendorf]

Closing, this was addressed in master and will be in 0.6.4.