Hi, we use OCFS2 as VM Storage and ran into a problem yesterday. The host reported the following: [Sun Sep 8 10:08:38 2013] ------------[ cut here ]------------ [Sun Sep 8 10:08:38 2013] kernel BUG at fs/ocfs2/extent_map.c:475! [Sun Sep 8 10:08:38 2013] invalid opcode: 0000 [#1] SMP [Sun Sep 8 10:08:38 2013] Modules linked in: vhost_net vhost macvtap macvlan tun ebtable_nat ebtables ocfs2_stack_o2cb bridge stp llc w83795 coretemp kvm_intel kvm drbd lru_cache dlm sctp libcrc32c ocfs2_dlm ocfs2_dlmfs ocfs2 ocfs2_stackglue ocfs2_nodemanager configfs e1000e ixgbe mdio [Sun Sep 8 10:08:38 2013] CPU: 18 PID: 19181 Comm: qemu-system-x86 Not tainted 3.11.0-rc7 #3 [Sun Sep 8 10:08:38 2013] Hardware name: Supermicro X8DT6/X8DT6, BIOS 2.0b 08/30/2011 [Sun Sep 8 10:08:38 2013] task: ffff882dc9dbbd80 ti: ffff882c71ba4000 task.ti: ffff882c71ba4000 [Sun Sep 8 10:08:38 2013] RIP: 0010:[<ffffffffa0157d2a>] [<ffffffffa0157d2a>] ocfs2_get_clusters_nocache.isra.11+0x48a/0x510 [ocfs2] [Sun Sep 8 10:08:38 2013] RSP: 0018:ffff882c71ba5728 EFLAGS: 00010297 [Sun Sep 8 10:08:38 2013] RAX: 0000000000000034 RBX: ffff882ce2bed030 RCX: 0000000000006021 [Sun Sep 8 10:08:38 2013] RDX: ffff882ce2bed380 RSI: 0000000000006021 RDI: ffff882ce2bed030 [Sun Sep 8 10:08:38 2013] RBP: ffff882c71ba5798 R08: 0000000000006022 R09: 0000000000000000 [Sun Sep 8 10:08:38 2013] R10: 000000000003fa1e R11: ffff882ce2bed000 R12: ffff882c71ba57d8 [Sun Sep 8 10:08:38 2013] R13: 0000000000006021 R14: ffff882e5c5a2000 R15: 0000000000000000 [Sun Sep 8 10:08:38 2013] FS: 00007fb8cffff700(0000) GS:ffff88303fcc0000(0000) knlGS:0000000000000000 [Sun Sep 8 10:08:38 2013] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b [Sun Sep 8 10:08:38 2013] CR2: 00000000c2e80000 CR3: 0000002b9d1c3000 CR4: 00000000000027e0 [Sun Sep 8 10:08:38 2013] DR0: 0000000000000001 DR1: 0000000000000002 DR2: 0000000000000001 [Sun Sep 8 10:08:38 2013] DR3: 000000000000000a DR6: 00000000ffff0ff0 DR7: 0000000000000400 [Sun Sep 8 10:08:38 2013] Stack: [Sun Sep 8 10:08:38 2013] ffff882c71ba5828 ffff882c71ba57d0 ffff882ce2bed000 ffff882c71ba57cc [Sun Sep 8 10:08:38 2013] ffff882cb2fb8a98 000000010060214a ffff882fd2d144e0 ffffffffa0163346 [Sun Sep 8 10:08:38 2013] 0000000000000002 0000000000006021 ffff882cb2fb8a98 ffff882c71ba58c4 [Sun Sep 8 10:08:38 2013] Call Trace: [Sun Sep 8 10:08:38 2013] [<ffffffffa0163346>] ? ocfs2_read_inode_block_full+0x36/0x60 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa01582b1>] ocfs2_get_clusters+0x221/0x390 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa0158472>] ocfs2_extent_map_get_blocks+0x52/0x180 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa013c4a1>] ocfs2_direct_IO_get_blocks+0x51/0x140 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffff8117858e>] do_blockdev_direct_IO+0x118e/0x1fa0 [Sun Sep 8 10:08:38 2013] [<ffffffff810f5b50>] ? pagevec_lookup_tag+0x20/0x30 [Sun Sep 8 10:08:38 2013] [<ffffffffa013c450>] ? ocfs2_dio_end_io+0x100/0x100 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffff811793f0>] __blockdev_direct_IO+0x50/0x60 [Sun Sep 8 10:08:38 2013] [<ffffffffa013c450>] ? ocfs2_dio_end_io+0x100/0x100 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa013c350>] ? ocfs2_direct_IO+0x80/0x80 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa013c343>] ocfs2_direct_IO+0x73/0x80 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa013c450>] ? ocfs2_dio_end_io+0x100/0x100 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffffa013c350>] ? ocfs2_direct_IO+0x80/0x80 [ocfs2] [Sun Sep 8 10:08:38 2013] [<ffffffff810eb974>] generic_file_aio_read+0x664/0x6c0 [Sun Sep 8 10:08:38 2013] [<ffffffffa015ae48>] ? ocfs2_should_update_atime+0x68/0x150 [ocfs2] [Sun Sep 8 10:08:39 2013] [<ffffffffa015160b>] ? __ocfs2_cluster_unlock.isra.40+0x9b/0xd0 [ocfs2] [Sun Sep 8 10:08:39 2013] [<ffffffffa015198a>] ? ocfs2_inode_unlock+0xaa/0x120 [ocfs2] [Sun Sep 8 10:08:39 2013] [<ffffffffa015aaed>] ocfs2_file_aio_read+0xad/0x3a0 [ocfs2] [Sun Sep 8 10:08:39 2013] [<ffffffff81029050>] ? native_smp_send_reschedule+0x40/0x60 [Sun Sep 8 10:08:39 2013] [<ffffffff8113ef2e>] do_sync_readv_writev+0x6e/0xa0 [Sun Sep 8 10:08:39 2013] [<ffffffff811402af>] do_readv_writev+0xcf/0x2b0 [Sun Sep 8 10:08:39 2013] [<ffffffff8114051d>] vfs_readv+0x2d/0x50 [Sun Sep 8 10:08:39 2013] [<ffffffff81140782>] SyS_preadv+0xc2/0xe0 [Sun Sep 8 10:08:39 2013] [<ffffffff816b5ed2>] system_call_fastpath+0x16/0x1b [Sun Sep 8 10:08:39 2013] Code: 00 02 00 00 49 c7 c0 30 27 1c a0 48 c7 c7 28 ef 1c a0 31 c0 e8 5e b9 54 e1 e9 1e fd ff ff 0f 1f 00 bb 01 00 00 00 e9 78 fe ff ff <0f> 0b 48 8b 55 b0 48 c7 c6 50 27 1c a0 bb e2 ff ff ff 4c 8b 47 [Sun Sep 8 10:08:39 2013] RIP [<ffffffffa0157d2a>] ocfs2_get_clusters_nocache.isra.11+0x48a/0x510 [ocfs2] [Sun Sep 8 10:08:39 2013] RSP <ffff882c71ba5728> [Sun Sep 8 10:08:39 2013] ---[ end trace aa04b37eb9874cc8 ]--- Unformatted: https://gist.github.com/David-Weber/c16e34cca09b1cfe9b10 Starting at this point, the guest had many read and write problems but not all of them failed. Restarting the process resolved the problem. The guest is a simple Windows Server 2003 which shouldn't do any fancy IO-operations. This only happens once so far, so I don't have a simple reproducer. Cheers, David