[Xen-users] Migration issues

[Russ Garrett <russ@xxxxxxxxxxxxx>]

I have an existing cluster of Xen machines to which I'm adding a new
machine. All machines are running Debian Squeeze, Xen 4.0.1, and Linux
2.6.32-5-xen-amd64, and for this test I'm using a Linux PV guest
running the same kernel.

The existing machines have Xeon L5420 (2.5GHz) CPUs, and the new
machine has Xeon L5520 (2.27GHz). CPU settings and DomU CPUID masks
are set so the /proc/cpuinfo "flags" line is identical on both
architectures.

When migrating a VM from the new machine to an old machine, I get
exactly the same behaviour as the bug [1]. The precise version of the
Debian kernel I'm running indicates that it *does* have the patch
referenced in [2], although given all the patching madness I'm only
99.9% sure of that.

When migrating in the opposite direction (from an old machine to the
new machine), I get the following oops (which I suspect might be an
unrelated bug...):

[   46.434736] BUG: unable to handle kernel paging request at 000000000a3d025c
[   46.434736] IP: [<ffffffff810327ea>] fill_pmd+0xe/0xec
[   46.434736] PGD fffffffffffff067 BAD
[   46.434736] Oops: 0000 [#1] SMP
[   46.434736] last sysfs file: /sys/devices/vbd-51713/block/xvda1/removable
[   46.434736] CPU 0
[   46.434736] Modules linked in: snd_pcm snd_timer snd soundcore
snd_page_alloc pcspkr evdev xfs exportfs dm_mod raid1 md_mod
xen_netfront xen_blkfront
[   46.434736] Pid: 793, comm: kstop/0 Not tainted 2.6.32-5-xen-amd64 #1
[   46.434736] RIP: e030:[<ffffffff810327ea>]  [<ffffffff810327ea>]
fill_pmd+0xe/0xec
[   46.434736] RSP: e02b:ffff88000299dd40  EFLAGS: 00010082
[   46.434736] RAX: 0000000000000ff8 RBX: ffffffffff57b000 RCX: ffffffffff57b000
[   46.434736] RDX: 00000000e7496063 RSI: ffffffffff57b000 RDI: ffffc7fffffffff8
[   46.434736] RBP: ffffc7fffffffff8 R08: 0000000000000000 R09: 7fffffffffffffff
[   46.434736] R10: ffffffff8100e63d R11: ffffc90000082000 R12: ffffffffff57b000
[   46.434736] R13: ffffe8fffffe6128 R14: ffff880002939530 R15: ffff880002939530
[   46.434736] FS:  00007fd322026700(0000) GS:ffff880003441000(0000)
knlGS:0000000000000000
[   46.434736] CS:  e033 DS: 0000 ES: 0000 CR0: 000000008005003b
[   46.434736] CR2: 000000000a3d025c CR3: 000000000df04000 CR4: 0000000000002660
[   46.434736] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[   46.434736] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
[   46.434736] Process kstop/0 (pid: 793, threadinfo ffff88000299c000,
task ffff880002939530)
[   46.434736] Stack:
[   46.434736]  ffffc90000082000 ffffffff8100c30f ffffffffff57b000
0000000000000884
[   46.434736] <0> 00000000e7496063 ffffffff810328e9 00000002fa6f3067
00000000e7496063
[   46.434736] <0> ffffffff81001000 00000000e7496063 ffffe8fffffe6120
ffffffff810369f3
[   46.434736] Call Trace:
[   46.434736]  [<ffffffff8100c30f>] ? __raw_callee_save_xen_pgd_val+0x11/0x1e
[   46.434736]  [<ffffffff810328e9>] ? set_pte_vaddr_pud+0x21/0x4b
[   46.434736]  [<ffffffff810369f3>] ? __native_set_fixmap+0x24/0x2c
[   46.434736]  [<ffffffff8100d091>] ? xen_set_fixmap+0xcc/0xfb
[   46.434736]  [<ffffffff8100c426>] ? pfn_to_mfn+0xe/0x22
[   46.434736]  [<ffffffff8100bf67>] ? xen_setup_shared_info+0x2d/0x56
[   46.434736]  [<ffffffff8100efdd>] ? xen_post_suspend+0x11/0x90
[   46.434736]  [<ffffffff811f396b>] ? xen_suspend+0x9a/0xc4
[   46.434736]  [<ffffffff8108bd2a>] ? stop_cpu+0x83/0xc6
[   46.434736]  [<ffffffff81062887>] ? worker_thread+0x188/0x21d
[   46.434736]  [<ffffffff8108bca7>] ? stop_cpu+0x0/0xc6
[   46.434736]  [<ffffffff81065eb2>] ? autoremove_wake_function+0x0/0x2e
[   46.434736]  [<ffffffff810626ff>] ? worker_thread+0x0/0x21d
[   46.434736]  [<ffffffff81065be5>] ? kthread+0x79/0x81
[   46.434736]  [<ffffffff81012baa>] ? child_rip+0xa/0x20
[   46.434736]  [<ffffffff81011d61>] ? int_ret_from_sys_call+0x7/0x1b
[   46.434736]  [<ffffffff8101251d>] ? retint_restore_args+0x5/0x6
[   46.434736]  [<ffffffff81012ba0>] ? child_rip+0x0/0x20
[   46.434736] Code: 00 00 48 ba 00 f0 ff ff ff 3f 00 00 41 5a 49 8d
04 04 48 21 d1 5b 5d 48 01 c8 41 5c c3 41 54 49 89 f4 55 48 89 fd 53
48 83 ec 10 <48> 83 3f 00 0f 85 95 00 00 00 e8 37 65 2c 00 48 89 c7 48
89 c3
[   46.434736] RIP  [<ffffffff810327ea>] fill_pmd+0xe/0xec
[   46.434736]  RSP <ffff88000299dd40>
[   46.434736] CR2: 000000000a3d025c
[   46.434736] ---[ end trace 5c8ab14266459bc5 ]---
[   46.434736] ------------[ cut here ]------------
[   46.434736] WARNING: at
/build/buildd-linux-2.6_2.6.32-31-amd64-vrfdM4/linux-2.6-2.6.32/debian/build/source_amd64_xen/kernel/time/timekeeping.c:249
ktime_get+0x22/0xb7()
[   46.434736] Modules linked in: snd_pcm snd_timer snd soundcore
snd_page_alloc pcspkr evdev xfs exportfs dm_mod raid1 md_mod
xen_netfront xen_blkfront
[   46.434736] Pid: 0, comm: swapper Tainted: G      D    2.6.32-5-xen-amd64 #1
[   46.434736] Call Trace:
[   46.434736]  [<ffffffff8106d48b>] ? ktime_get+0x22/0xb7
[   46.434736]  [<ffffffff8106d48b>] ? ktime_get+0x22/0xb7
[   46.434736]  [<ffffffff8104ef10>] ? warn_slowpath_common+0x77/0xa3
[   46.434736]  [<ffffffff8106d48b>] ? ktime_get+0x22/0xb7
[   46.434736]  [<ffffffff810719c1>] ? tick_nohz_stop_sched_tick+0x6f/0x335
[   46.434736]  [<ffffffff81010e82>] ? cpu_idle+0x73/0xda
[   46.434736]  [<ffffffff81509cdd>] ? start_kernel+0x3dc/0x3e8
[   46.434736]  [<ffffffff8150bc93>] ? xen_start_kernel+0x586/0x58a
[   46.434736] ---[ end trace 5c8ab14266459bc6 ]---

Cheers,

[1] https://bugzilla.redhat.com/show_bug.cgi?id=663755
[2] https://bugzilla.redhat.com/show_bug.cgi?id=663755#c20

-- 
Russ Garrett
russ@xxxxxxxxxxxxx

_______________________________________________
Xen-users mailing list
Xen-users@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-users