[sheepdog-users] 答复: sheepdog-users Digest, Vol 14, Issue 40

George Y. Hu huyuanyuan at gamutsoft.com
Tue Jul 2 04:59:05 CEST 2013 

Dears,

I installed corosync(1.4.6)+sheepdog(0.6.0) on two Centos6, with the
following configuration of corosync.conf

-----------------------------------------
compatibility: whitetank

totem {
        version: 2
        secauth: off
        threads: 0
        interface {
                ringnumber: 0
                bindnetaddr: 10.86.213.251 (252 is another)
                mcastaddr: 226.94.1.1
                mcastport: 5405
                ttl: 1
        }
}

logging {
        fileline: off
        to_stderr: no
        to_logfile: yes
        logfile: /var/log/cluster/corosync.log
        to_syslog: yes
        debug: off
        timestamp: on
        logger_subsys {
                subsys: AMF
                debug: off
        }
}
--------------------------------------------

When I start sheepdog service by "sheep /var/lib/sheep", it seems two nodes
are not connected since I can see only one node in "collie node list",
M   Id   Host:Port         V-Nodes       Zone
-    0   10.86.213.251:7000     64  -69904886

Iptables has been disabled but the problem remains.
Will somebody help me on that?


Best Regards,

George Y. Hu


Send sheepdog-users mailing list submissions to
	sheepdog-users at lists.wpkg.org

To subscribe or unsubscribe via the World Wide Web, visit
	http://lists.wpkg.org/mailman/listinfo/sheepdog-users
or, via email, send a message with subject or body 'help' to
	sheepdog-users-request at lists.wpkg.org

You can reach the person managing the list at
	sheepdog-users-owner at lists.wpkg.org

When replying, please edit your Subject line so it is more specific
than "Re: Contents of sheepdog-users digest..."


Today's Topics:

   1. Re: Problem with snapshots made with qemu-img (Liu Yuan)
   2. Crash khugepaged (Valerio Pachera)
   3. Re: Crash khugepaged (Valerio Pachera)
   4. Re: cluster format during recovery (MORITA Kazutaka)


----------------------------------------------------------------------

Message: 1
Date: Fri, 28 Jun 2013 18:02:59 +0800
From: Liu Yuan <namei.unix at gmail.com>
To: "Ing. Luca Lazzeroni - Trend Servizi Srl" <luca at gvnet.it>
Cc: "sheepdog-users at lists.wpkg.org" <sheepdog-users at lists.wpkg.org>
Subject: Re: [sheepdog-users] Problem with snapshots made with
	qemu-img
Message-ID: <20130628100259.GC13194 at ubuntu-precise>
Content-Type: text/plain; charset=utf-8

On Fri, Jun 28, 2013 at 10:07:47AM +0200, Ing. Luca Lazzeroni - Trend
Servizi Srl wrote:
> Hi,
> if I make a snapshot of a running VM using:
> 
> qemu-img snapshot -c Pippo Pluto.raw
> 
> snapshot is created on all nodes, but its tag is updated on all nodes
except the one running the VM.
> On other nodes I can see, via "collie vdi list" the snapshot tag updated
correctly, but on the node running the VM I see 2 VDI with the same name,
different ID and empty Tag.

Seems that recent qemu-img need fixes, we didn't test snapshot with qemu-img
with our functonal tests. We should though.

> 
> If I create the snapshot via "collie vdi snapshot", everything works fine
and tag is propagated to all nodes; but I don't know if creating a snapshot
with collie of a running VM with writeback cache enabled is a good idea in
terms of data integrity?

No problem, snapshot operation will
1 flush the cache first
2 mark the vdi as readonly

If there is, it is a bug that should be fixed.

Thanks,
Yuan


------------------------------

Message: 2
Date: Fri, 28 Jun 2013 17:40:26 +0200
From: Valerio Pachera <sirio81 at gmail.com>
To: Lista sheepdog user <sheepdog-users at lists.wpkg.org>
Subject: [sheepdog-users] Crash khugepaged
Message-ID:
	<CAHS0cb-KqoS6wWt_gT+bSQ56KS7Z5iA4yOSpX5zQsoGPX0WV=Q at mail.gmail.com>
Content-Type: text/plain; charset=UTF-8

What do you think about this?

Jun 28 16:34:20 sheepdog004 kernel: [103658.606691] khugepaged      D
ffff88021f393780     0    32      2 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.606696]  ffff880213793750
0000000000000046 ffffffff00000000 ffff880216566f60
Jun 28 16:34:20 sheepdog004 kernel: [103658.606701]  0000000000013780
ffff880213795fd8 ffff880213795fd8 ffff880213793750
Jun 28 16:34:20 sheepdog004 kernel: [103658.606713]  ffff880213795730
0000000113795730 ffff88021657fe50 ffff88021f393fd0
Jun 28 16:34:20 sheepdog004 kernel: [103658.606718] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.606727]
[<ffffffff810b47b3>] ? lock_page+0x20/0x20
Jun 28 16:34:20 sheepdog004 kernel: [103658.606732]
[<ffffffff8134da71>] ? io_schedule+0x59/0x71
Jun 28 16:34:20 sheepdog004 kernel: [103658.606737]
[<ffffffff810b47b9>] ? sleep_on_page+0x6/0xa
Jun 28 16:34:20 sheepdog004 kernel: [103658.606740]
[<ffffffff8134deb4>] ? __wait_on_bit+0x3e/0x71
Jun 28 16:34:20 sheepdog004 kernel: [103658.606744]
[<ffffffff810b48f5>] ? wait_on_page_bit+0x6e/0x73
Jun 28 16:34:20 sheepdog004 kernel: [103658.606751]
[<ffffffff8105fb09>] ? autoremove_wake_function+0x2a/0x2a
Jun 28 16:34:20 sheepdog004 kernel: [103658.606756]
[<ffffffff810c2850>] ? shrink_page_list+0x166/0x73f
Jun 28 16:34:20 sheepdog004 kernel: [103658.606761]
[<ffffffff810c9cfa>] ? zone_page_state_add+0x14/0x23
Jun 28 16:34:20 sheepdog004 kernel: [103658.606765]
[<ffffffff810c0e13>] ? update_isolated_counts+0x13b/0x15a
Jun 28 16:34:20 sheepdog004 kernel: [103658.606769]
[<ffffffff810c32c4>] ? shrink_inactive_list+0x2cd/0x3f0
Jun 28 16:34:20 sheepdog004 kernel: [103658.606774]
[<ffffffff810be232>] ? __lru_cache_add+0x2b/0x51
Jun 28 16:34:20 sheepdog004 kernel: [103658.606778]
[<ffffffff810c3a89>] ? shrink_zone+0x3c0/0x4e6
Jun 28 16:34:20 sheepdog004 kernel: [103658.606783]
[<ffffffff810c3fa7>] ? do_try_to_free_pages+0x1cc/0x41c
Jun 28 16:34:20 sheepdog004 kernel: [103658.606787]
[<ffffffff810c4462>] ? try_to_free_pages+0xa9/0xe9
Jun 28 16:34:20 sheepdog004 kernel: [103658.606791]
[<ffffffff810364e8>] ? should_resched+0x5/0x23
Jun 28 16:34:20 sheepdog004 kernel: [103658.606796]
[<ffffffff810bb3ee>] ? __alloc_pages_nodemask+0x4ed/0x7aa
Jun 28 16:34:20 sheepdog004 kernel: [103658.606801]
[<ffffffff8100d69f>] ? __switch_to+0x133/0x258
Jun 28 16:34:20 sheepdog004 kernel: [103658.606806]
[<ffffffff8134eb77>] ? _raw_spin_unlock_irqrestore+0xe/0xf
Jun 28 16:34:20 sheepdog004 kernel: [103658.606811]
[<ffffffff810e5f05>] ? alloc_pages_vma+0x12d/0x136
Jun 28 16:34:20 sheepdog004 kernel: [103658.606815]
[<ffffffff810ce1c5>] ? pte_pfn+0x5/0xe
Jun 28 16:34:20 sheepdog004 kernel: [103658.606819]
[<ffffffff810ef9bd>] ? khugepaged+0x4dc/0xef3
Jun 28 16:34:20 sheepdog004 kernel: [103658.606823]
[<ffffffff8100d69f>] ? __switch_to+0x133/0x258
Jun 28 16:34:20 sheepdog004 kernel: [103658.606828]
[<ffffffff8105fadf>] ? add_wait_queue+0x3c/0x3c
Jun 28 16:34:20 sheepdog004 kernel: [103658.606833]
[<ffffffff810ef4e1>] ? add_mm_counter.constprop.28+0x9/0x9
Jun 28 16:34:20 sheepdog004 kernel: [103658.606837]
[<ffffffff8105f48d>] ? kthread+0x76/0x7e
Jun 28 16:34:20 sheepdog004 kernel: [103658.606842]
[<ffffffff81355cb4>] ? kernel_thread_helper+0x4/0x10
Jun 28 16:34:20 sheepdog004 kernel: [103658.606847]
[<ffffffff8105f417>] ? kthread_worker_fn+0x139/0x139
Jun 28 16:34:20 sheepdog004 kernel: [103658.606851]
[<ffffffff81355cb0>] ? gs_change+0x13/0x13
Jun 28 16:34:20 sheepdog004 kernel: [103658.606983] sheep           D
ffff88021f393780     0 30859      1 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.606987]  ffff880101d48730
0000000000000082 0000000000000000 ffff880216566f60
Jun 28 16:34:20 sheepdog004 kernel: [103658.606992]  0000000000013780
ffff8802141dffd8 ffff8802141dffd8 ffff880101d48730
Jun 28 16:34:20 sheepdog004 kernel: [103658.606997]  ffffea00048c4b20
0000000105019098 ffffea0004fdaaa8 ffff880214677be0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607001] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.607005]
[<ffffffff8134eac4>] ? rwsem_down_failed_common+0xe0/0x114
Jun 28 16:34:20 sheepdog004 kernel: [103658.607011]
[<ffffffff811b3af3>] ? call_rwsem_down_write_failed+0x13/0x20
Jun 28 16:34:20 sheepdog004 kernel: [103658.607015]
[<ffffffff8134e431>] ? down_write+0x25/0x27
Jun 28 16:34:20 sheepdog004 kernel: [103658.607019]
[<ffffffff810d543d>] ? sys_munmap+0x2e/0x52
Jun 28 16:34:20 sheepdog004 kernel: [103658.607023]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607135] tar             D
ffff88021f293780     0 14370  13938 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.607139]  ffff88021472ae60
0000000000000086 ffffffff00000000 ffff8802165160c0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607144]  0000000000013780
ffff880128b77fd8 ffff880128b77fd8 ffff88021472ae60
Jun 28 16:34:20 sheepdog004 kernel: [103658.607148]  ffffffff8101360a
00000001810660a1 ffff880213ff3f30 ffff88021f293fd0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607001] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.607005]
[<ffffffff8134eac4>] ? rwsem_down_failed_common+0xe0/0x114
Jun 28 16:34:20 sheepdog004 kernel: [103658.607011]
[<ffffffff811b3af3>] ? call_rwsem_down_write_failed+0x13/0x20
Jun 28 16:34:20 sheepdog004 kernel: [103658.607015]
[<ffffffff8134e431>] ? down_write+0x25/0x27
Jun 28 16:34:20 sheepdog004 kernel: [103658.607019]
[<ffffffff810d543d>] ? sys_munmap+0x2e/0x52
Jun 28 16:34:20 sheepdog004 kernel: [103658.607023]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607135] tar             D
ffff88021f293780     0 14370  13938 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.607139]  ffff88021472ae60
0000000000000086 ffffffff00000000 ffff8802165160c0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607144]  0000000000013780
ffff880128b77fd8 ffff880128b77fd8 ffff88021472ae60
Jun 28 16:34:20 sheepdog004 kernel: [103658.607148]  ffffffff8101360a
00000001810660a1 ffff880213ff3f30 ffff88021f293fd0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607153] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.607157]
[<ffffffff8101360a>] ? read_tsc+0x5/0x14
Jun 28 16:34:20 sheepdog004 kernel: [103658.607161]
[<ffffffff810b47b3>] ? lock_page+0x20/0x20
Jun 28 16:34:20 sheepdog004 kernel: [103658.607165]
[<ffffffff8134da71>] ? io_schedule+0x59/0x71
Jun 28 16:34:20 sheepdog004 kernel: [103658.607169]
[<ffffffff810b47b9>] ? sleep_on_page+0x6/0xa
Jun 28 16:34:20 sheepdog004 kernel: [103658.607172]
[<ffffffff8134deb4>] ? __wait_on_bit+0x3e/0x71
Jun 28 16:34:20 sheepdog004 kernel: [103658.607176]
[<ffffffff810b48f5>] ? wait_on_page_bit+0x6e/0x73
Jun 28 16:34:20 sheepdog004 kernel: [103658.607181]
[<ffffffff8105fb09>] ? autoremove_wake_function+0x2a/0x2a
Jun 28 16:34:20 sheepdog004 kernel: [103658.607186]
[<ffffffff810b49cd>] ? filemap_fdatawait_range+0x74/0x139
Jun 28 16:34:20 sheepdog004 kernel: [103658.607191]
[<ffffffff810b6181>] ? filemap_write_and_wait+0x24/0x30
Jun 28 16:34:20 sheepdog004 kernel: [103658.607205]
[<ffffffffa053ac73>] ? nfs_getattr+0x32/0xac [nfs]
Jun 28 16:34:20 sheepdog004 kernel: [103658.607211]
[<ffffffff810fda17>] ? vfs_fstat+0x30/0x4e
Jun 28 16:34:20 sheepdog004 kernel: [103658.607214]
[<ffffffff810fdb49>] ? sys_newfstat+0x12/0x2b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607218]
[<ffffffff810fa376>] ? vfs_write+0xbb/0xe9
Jun 28 16:34:20 sheepdog004 kernel: [103658.607221]
[<ffffffff810fa554>] ? sys_write+0x5f/0x6b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607225]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607335] pgrep           D
ffff88021f293780     0 30870  30869 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.607339]  ffff880133e7c730
0000000000000086 0000000100000000 ffff8802165160c0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607344]  0000000000013780
ffff8801340f5fd8 ffff8801340f5fd8 ffff880133e7c730
Jun 28 16:34:20 sheepdog004 kernel: [103658.607349]  0000000000000020
000000011f5fcc08 0000000000000002 ffff880214677be0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607353] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.607357]
[<ffffffff8134eac4>] ? rwsem_down_failed_common+0xe0/0x114
Jun 28 16:34:20 sheepdog004 kernel: [103658.607361]
[<ffffffff811b3ac4>] ? call_rwsem_down_read_failed+0x14/0x30
Jun 28 16:34:20 sheepdog004 kernel: [103658.607365]
[<ffffffff8134e44a>] ? down_read+0x17/0x19
Jun 28 16:34:20 sheepdog004 kernel: [103658.607369]
[<ffffffff810d1a94>] ? __access_remote_vm+0x3a/0x1c1
Jun 28 16:34:20 sheepdog004 kernel: [103658.607374]
[<ffffffff810d2acb>] ? access_process_vm+0x48/0x65
Jun 28 16:34:20 sheepdog004 kernel: [103658.607378]
[<ffffffff81140852>] ? proc_pid_cmdline+0x63/0xf0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607382]
[<ffffffff81141a58>] ? proc_info_read+0x5b/0xb8
Jun 28 16:34:20 sheepdog004 kernel: [103658.607386]
[<ffffffff810fa443>] ? vfs_read+0x9f/0xe6
Jun 28 16:34:20 sheepdog004 kernel: [103658.607390]
[<ffffffff810fa4cf>] ? sys_read+0x45/0x6b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607393]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607503] pgrep           D
ffff88021f293780     0 30926  30925 0x00000000
Jun 28 16:34:20 sheepdog004 kernel: [103658.607507]  ffff880212ed2e20
0000000000000086 0000000100000000 ffff8802165160c0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607512]  0000000000013780
ffff880132047fd8 ffff880132047fd8 ffff880212ed2e20
Jun 28 16:34:20 sheepdog004 kernel: [103658.607517]  0000000000000020
000000011f5fcc08 0000000000000002 ffff880214677be0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607521] Call Trace:
Jun 28 16:34:20 sheepdog004 kernel: [103658.607525]
[<ffffffff8134eac4>] ? rwsem_down_failed_common+0xe0/0x114
Jun 28 16:34:20 sheepdog004 kernel: [103658.607529]
[<ffffffff811b3ac4>] ? call_rwsem_down_read_failed+0x14/0x30
Jun 28 16:34:20 sheepdog004 kernel: [103658.607533]
[<ffffffff8134e44a>] ? down_read+0x17/0x19
Jun 28 16:34:20 sheepdog004 kernel: [103658.607537]
[<ffffffff810d1a94>] ? __access_remote_vm+0x3a/0x1c1
Jun 28 16:34:20 sheepdog004 kernel: [103658.607541]
[<ffffffff810d2acb>] ? access_process_vm+0x48/0x65
Jun 28 16:34:20 sheepdog004 kernel: [103658.607545]
[<ffffffff81140852>] ? proc_pid_cmdline+0x63/0xf0
Jun 28 16:34:20 sheepdog004 kernel: [103658.607548]
[<ffffffff81141a58>] ? proc_info_read+0x5b/0xb8
Jun 28 16:34:20 sheepdog004 kernel: [103658.607552]
[<ffffffff810fa443>] ? vfs_read+0x9f/0xe6
Jun 28 16:34:20 sheepdog004 kernel: [103658.607556]
[<ffffffff810fa4cf>] ? sys_read+0x45/0x6b
Jun 28 16:34:20 sheepdog004 kernel: [103658.607559]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:36:20 sheepdog004 kernel: [103778.581543] Call Trace:
Jun 28 16:36:20 sheepdog004 kernel: [103778.581552]
[<ffffffff810b47b3>] ? lock_page+0x20/0x20
Jun 28 16:36:20 sheepdog004 kernel: [103778.581557]
[<ffffffff8134da71>] ? io_schedule+0x59/0x71
Jun 28 16:36:20 sheepdog004 kernel: [103778.581561]
[<ffffffff810b47b9>] ? sleep_on_page+0x6/0xa
Jun 28 16:36:20 sheepdog004 kernel: [103778.581565]
[<ffffffff8134deb4>] ? __wait_on_bit+0x3e/0x71
Jun 28 16:36:20 sheepdog004 kernel: [103778.581569]
[<ffffffff810b48f5>] ? wait_on_page_bit+0x6e/0x73
Jun 28 16:36:20 sheepdog004 kernel: [103778.581575]
[<ffffffff8105fb09>] ? autoremove_wake_function+0x2a/0x2a
Jun 28 16:36:20 sheepdog004 kernel: [103778.581581]
[<ffffffff810c2850>] ? shrink_page_list+0x166/0x73f
Jun 28 16:36:20 sheepdog004 kernel: [103778.581586]
[<ffffffff810c9cfa>] ? zone_page_state_add+0x14/0x23
Jun 28 16:36:20 sheepdog004 kernel: [103778.581591]
[<ffffffff810c0e13>] ? update_isolated_counts+0x13b/0x15a
Jun 28 16:36:20 sheepdog004 kernel: [103778.581595]
[<ffffffff810c32c4>] ? shrink_inactive_list+0x2cd/0x3f0
Jun 28 16:36:20 sheepdog004 kernel: [103778.581600]
[<ffffffff810be232>] ? __lru_cache_add+0x2b/0x51
Jun 28 16:36:20 sheepdog004 kernel: [103778.581604]
[<ffffffff810c3a89>] ? shrink_zone+0x3c0/0x4e6
Jun 28 16:36:20 sheepdog004 kernel: [103778.581608]
[<ffffffff810c3fa7>] ? do_try_to_free_pages+0x1cc/0x41c
Jun 28 16:36:20 sheepdog004 kernel: [103778.581612]
[<ffffffff810c4462>] ? try_to_free_pages+0xa9/0xe9
Jun 28 16:36:20 sheepdog004 kernel: [103778.581616]
[<ffffffff810364e8>] ? should_resched+0x5/0x23
Jun 28 16:36:20 sheepdog004 kernel: [103778.581621]
[<ffffffff810bb3ee>] ? __alloc_pages_nodemask+0x4ed/0x7aa
Jun 28 16:36:20 sheepdog004 kernel: [103778.581626]
[<ffffffff8100d69f>] ? __switch_to+0x133/0x258
Jun 28 16:36:20 sheepdog004 kernel: [103778.581631]
[<ffffffff8134eb77>] ? _raw_spin_unlock_irqrestore+0xe/0xf
Jun 28 16:36:20 sheepdog004 kernel: [103778.581636]
[<ffffffff810e5f05>] ? alloc_pages_vma+0x12d/0x136
Jun 28 16:36:20 sheepdog004 kernel: [103778.581640]
[<ffffffff810ce1c5>] ? pte_pfn+0x5/0xe
Jun 28 16:36:20 sheepdog004 kernel: [103778.581645]
[<ffffffff810ef9bd>] ? khugepaged+0x4dc/0xef3
Jun 28 16:36:20 sheepdog004 kernel: [103778.581649]
[<ffffffff8100d69f>] ? __switch_to+0x133/0x258
Jun 28 16:36:20 sheepdog004 kernel: [103778.581654]
[<ffffffff8105fadf>] ? add_wait_queue+0x3c/0x3c
Jun 28 16:36:20 sheepdog004 kernel: [103778.581658]
[<ffffffff810ef4e1>] ? add_mm_counter.constprop.28+0x9/0x9
Jun 28 16:36:20 sheepdog004 kernel: [103778.581662]
[<ffffffff8105f48d>] ? kthread+0x76/0x7e
Jun 28 16:36:20 sheepdog004 kernel: [103778.581667]
[<ffffffff81355cb4>] ? kernel_thread_helper+0x4/0x10
Jun 28 16:36:20 sheepdog004 kernel: [103778.581671]
[<ffffffff8105f417>] ? kthread_worker_fn+0x139/0x139
Jun 28 16:36:20 sheepdog004 kernel: [103778.581675]
[<ffffffff81355cb0>] ? gs_change+0x13/0x13
Jun 28 16:36:20 sheepdog004 kernel: [103778.581808] sheep           D
ffff88021f393780     0 30859      1 0x00000000
Jun 28 16:36:20 sheepdog004 kernel: [103778.581813]  ffff880101d48730
0000000000000082 0000000000000000 ffff880216566f60
Jun 28 16:36:20 sheepdog004 kernel: [103778.581817]  0000000000013780
ffff8802141dffd8 ffff8802141dffd8 ffff880101d48730
Jun 28 16:36:20 sheepdog004 kernel: [103778.581822]  ffffea00048c4b20
0000000105019098 ffffea0004fdaaa8 ffff880214677be0
Jun 28 16:36:20 sheepdog004 kernel: [103778.581827] Call Trace:
Jun 28 16:36:20 sheepdog004 kernel: [103778.581831]
[<ffffffff8134eac4>] ? rwsem_down_failed_common+0xe0/0x114
Jun 28 16:36:20 sheepdog004 kernel: [103778.581842]
[<ffffffff811b3af3>] ? call_rwsem_down_write_failed+0x13/0x20
Jun 28 16:36:20 sheepdog004 kernel: [103778.581846]
[<ffffffff8134e431>] ? down_write+0x25/0x27
Jun 28 16:36:20 sheepdog004 kernel: [103778.581850]
[<ffffffff810d543d>] ? sys_munmap+0x2e/0x52
Jun 28 16:36:20 sheepdog004 kernel: [103778.581854]
[<ffffffff81353b52>] ? system_call_fastpath+0x16/0x1b
Jun 28 16:36:20 sheepdog004 kernel: [103778.581969] tar             D
ffff88021f293780     0 14370  13938 0x00000000
Jun 28 16:36:20 sheepdog004 kernel: [103778.581974]  ffff88021472ae60
0000000000000086 ffffffff00000000 ffff8802165160c0
Jun 28 16:36:20 sheepdog004 kernel: [103778.581978]  0000000000013780
ffff880128b77fd8 ffff880128b77fd8 ffff88021472ae60
Jun 28 16:36:20 sheepdog004 kernel: [103778.581983]  ffffffff8101360a
00000001810660a1 ffff880213ff3f30 ffff88021f293fd0
Jun 28 16:36:20 sheepdog004 kernel: [103778.581988] Call Trace:
Jun 28 16:36:20 sheepdog004 kernel: [103778.581992]
[<ffffffff8101360a>] ? read_tsc+0x5/0x14
Jun 28 16:36:20 sheepdog004 kernel: [103778.581996]
[<ffffffff810b47b3>] ? lock_page+0x20/0x20
Jun 28 16:36:20 sheepdog004 kernel: [103778.581999]
[<ffffffff8134da71>] ? io_schedule+0x59/0x71
Jun 28 16:36:20 sheepdog004 kernel: [103778.582003]
[<ffffffff810b47b9>] ? sleep_on_page+0x6/0xa
....

Host with 8G of ram.
The host was exporting also a nfs folder.
Guest was mounting this folder.
Guest for decompressing a big tar.gz (77G).


------------------------------

Message: 3
Date: Fri, 28 Jun 2013 18:02:37 +0200
From: Valerio Pachera <sirio81 at gmail.com>
To: Lista sheepdog user <sheepdog-users at lists.wpkg.org>
Subject: Re: [sheepdog-users] Crash khugepaged
Message-ID:
	<CAHS0cb8TogSOD2pGE+TsScm+o=g1kEXGdUNoFWWF-xYoVgfwog at mail.gmail.com>
Content-Type: text/plain; charset=UTF-8

2013/6/28 Valerio Pachera <sirio81 at gmail.com>:
> What do you think about this?

The crash was host side.
It was difficult to interact with the host because pgrep, atop, pa
aux, were freezing.
'top' and 'kill' were working.
I had to kill -9 the guests.
I've been able to reboot the host (and first shutdown the cluster).
Collie node list was showing the host still inside the cluster.

I wonder if the crash may be related to excessive network traffic on
the nic, or it's related to the use of transparent huge pages.
I set back the default value (madvide) but I'm not going to repeat the
decompression via nfs today.


------------------------------

Message: 4
Date: Sat, 29 Jun 2013 12:50:06 +0900
From: MORITA Kazutaka <morita.kazutaka at gmail.com>
To: Valerio Pachera <sirio81 at gmail.com>
Cc: Lista sheepdog user <sheepdog-users at lists.wpkg.org>
Subject: Re: [sheepdog-users] cluster format during recovery
Message-ID: <m27ghd60ep.wl%morita.kazutaka at gmail.com>
Content-Type: text/plain; charset=US-ASCII

At Thu, 27 Jun 2013 15:45:36 +0200,
Valerio Pachera wrote:
> 
> This is an unusual thing.
> It's useful for testing purpose only:
> 
> What happens if cluster format is run during a recovery?

Probably, the recovery process will print a lot of error messages
after cluster format since it cannot find any objects to be recovered.

Thanks,

Kazutaka


------------------------------

_______________________________________________
sheepdog-users mailing list
sheepdog-users at lists.wpkg.org
http://lists.wpkg.org/mailman/listinfo/sheepdog-users


End of sheepdog-users Digest, Vol 14, Issue 40
**********************************************

More information about the sheepdog-users mailing list