__write_lock_failed

__write_lock_failed

[stefanx]

Hello,

some of my virtual machines (kvm/qemu) sometimes crash (xfs, Ubuntu 10.04, 2.6.32-31-x86_64).
I think it happens while taking LVM-snapshots from the XFS-Filesystem of that machines:

 BUG: soft lockup - CPU#7 stuck for 61s! [kswapd0:84]
 Modules linked in: reiserfs ipt_ULOG ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ipt_REJECT xt_tcpudp iptable_filter ip_tables x_tables kvm_intel kvm xfs exportfs bridge stp ppdev fbcon tileblit font bitblit softcursor parport_pc psmouse serio_raw i5000_edac bnx2 vga16fb vgastate edac_core i5k_amb shpchp lp parport raid10 raid456 async_pq async_xor xor async_memcpy async_raid6_recov raid6_pq async_tx raid1 raid0 floppy multipath ahci e1000e linear
 CPU 7:
 Modules linked in: reiserfs ipt_ULOG ipt_MASQUERADE iptable_nat nf_nat nf_conntrack_ipv4 nf_defrag_ipv4 xt_state nf_conntrack ipt_REJECT xt_tcpudp iptable_filter ip_tables x_tables kvm_intel kvm xfs exportfs bridge stp ppdev fbcon tileblit font bitblit softcursor parport_pc psmouse serio_raw i5000_edac bnx2 vga16fb vgastate edac_core i5k_amb shpchp lp parport raid10 raid456 async_pq async_xor xor async_memcpy async_raid6_recov raid6_pq async_tx raid1 raid0 floppy multipath ahci e1000e linear
 Pid: 84, comm: kswapd0 Not tainted 2.6.32-27-server #49-Ubuntu PRIMERGY TX200 S4
 RIP: 0010:[]  [] __write_lock_failed+0x9/0x20
 RSP: 0018:ffff88062237dbc8  EFLAGS: 00000206
 RAX: 0000000000000000 RBX: ffff88062237dbd0 RCX: 0000000000000000
 RDX: ffffffffa01c3a00 RSI: 0000000000000000 RDI: ffff8806190c6048
 RBP: ffffffff81012cae R08: 0000000000000000 R09: 0000000000000001
 R10: 0000000000000000 R11: 000000000000007c R12: ffff88061ae424d0
 R13: ffff8802b2c18c00 R14: ffff88061ae424d0 R15: 0000000000000202
 FS:  0000000000000000(0000) GS:ffff880016fc0000(0000) knlGS:0000000000000000
 CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
 CR2: 0000000000000000 CR3: 0000000001001000 CR4: 00000000000026e0
 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
 DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
 Call Trace:
  [] ? _write_lock+0x17/0x20
  [] ? xfs_inode_ag_walk+0x68/0x140 [xfs]
  [] ? xfs_reclaim_inode+0x0/0x120 [xfs]
  [] ? xfs_inode_ag_iterator+0x99/0xe0 [xfs]
  [] ? xfs_reclaim_inode+0x0/0x120 [xfs]
  [] ? xfs_reclaim_inode_shrink+0xfc/0x140 [xfs]
  [] ? shrink_slab+0x125/0x190
  [] ? balance_pgdat+0x526/0x6d0
  [] ? isolate_pages_global+0x0/0x50
  [] ? kswapd+0xfe/0x150
  [] ? autoremove_wake_function+0x0/0x40
  [] ? kswapd+0x0/0x150
  [] ? kthread+0x96/0xa0
  [] ? child_rip+0xa/0x20


This link seems to describe the same problem with Xen:

http://lists.xensource.com/archives/html/xen-users/2010-04/msg01135.html


Any ideas ?

Thanks

Stefan

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: __write_lock_failed

[Christoph Hellwig]

On Fri, May 06, 2011 at 02:11:18PM +0200, stefanx wrote:
> 
> Hello,
> 
> some of my virtual machines (kvm/qemu) sometimes crash (xfs, Ubuntu 10.04, 2.6.32-31-x86_64).
> I think it happens while taking LVM-snapshots from the XFS-Filesystem of that machines:

This area has gotten a lot of changes since 2.6.32, including removing
the lock that's causing softlockups for you.  I don't think there's much
of a chance for help with a kernel that old as people's minds have moved
on to much more recent codebases.

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs