[Ocfs2-users] another fencing question

Sunil Mushran sunil.mushran at oracle.com
Fri Feb 5 11:16:00 PST 2010 

iozone is a fs benchmark. That won't show storage issues.

Angelo McComis wrote:
> Someone around the cubes mentioned using a tool called iozone to test
> disk i/o on these luns that are ocfs2.  With the collective knowledge
> here, is this ok to do? Smart, not smart, useful, not useful?   These
> stalled I/Os are hurting me too, but on storage that's not EMC... so
> this is also relevant to my current pains.
>
> On Fri, Feb 5, 2010 at 3:16 AM, Mailing List SVR
> <lists at svrinformatica.it> wrote:
>   
>> EMC installed statistics collection tool but until today nothing strange was
>> observed and the hangs continue ...
>>
>> I was forced to install a watchdog and reboot the servers when touch <file>
>> doesn't complete after 30 seconds,
>>
>> the logs posted in the previous emails doesn't suggest any problem on ocfs2
>> side?
>>
>> thanks
>> Nicola
>>
>> In data mercoledì 27 gennaio 2010 17:14:39, Mailing List SVR ha scritto:
>> : > Please note that at 15:29:30 I did a manual reboot on the node1, so the
>>     
>>>  hang on the node2 was from 14:59:34 to 15:29:30,
>>>
>>> regards
>>> Nicola
>>>
>>> In data mercoledì 27 gennaio 2010 17:03:07, Mailing List SVR ha scritto:
>>> : > Hi Sunil,
>>>       
>>>> another hang here:
>>>>
>>>> - node2 (making heavy I/O operation at that time) was up, ping to it was
>>>> ok (following your suggestion I started a ping from node1 to node2 some
>>>> days ago and no packet was loss, so the switch are ok) but I cannot login
>>>> via ssh or from the console (hang)
>>>> - node1 was up with low load but was unable to write to the ocfs2
>>>> partition (a simple touch <file> hang).
>>>> - after rebooting node1, node2 restarted to work and I was able to access
>>>>  to it,
>>>>
>>>> I'm investigating with EMC for hardware problems too but this seems to me
>>>>  an ocfs2 problem.
>>>>
>>>> Here are some logs:
>>>>
>>>> node1:
>>>>
>>>> Jan 27 14:59:33 nvr1-rc kernel: o2net: connection to node
>>>> nvr2-rc.minint.it (num 0) at 1.1.1.6:7777 has been idle for 35.0 seconds,
>>>> shutting it down. Jan 27 14:59:33 nvr1-rc kernel:
>>>> (0,7):o2net_idle_timer:1503 here are some times that might help debug the
>>>> situation: (tmr 1264600738.986852 now 1264600773.986647 dr
>>>> 1264600738.989741 adv
>>>>  1264600738.989753:1264600738.989754 func (75554989:507)
>>>>  1264600736.951519:1264600736.951521)
>>>> Jan 27 14:59:33 nvr1-rc kernel: o2net: no longer connected to node nvr2-
>>>> rc.minint.it (num 0) at 1.1.1.6:7777
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -112
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -112
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_purge_lockres:206 ERROR:
>>>>  status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_send_proxy_ast_msg:458
>>>> ERROR: status = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: (4025,5):dlm_flush_asts:600 ERROR: status
>>>> = -107
>>>> Jan 27 14:59:33 nvr1-rc kernel: o2net: connected to node
>>>> nvr2-rc.minint.it (num 0) at 1.1.1.6:7777
>>>>
>>>> node2:
>>>>
>>>> Jan 27 14:59:34 nvr2-rc kernel: o2net: no longer connected to node nvr1-
>>>> rc.minint.it (num 1) at 1.1.1.5:7777
>>>> Jan 27 14:59:34 nvr2-rc kernel: (4000,4):dlm_drop_lockres_ref:2211 ERROR:
>>>> status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel:
>>>> (6208,5):dlm_send_remote_unlock_request:359 ERROR: status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel: (3344,3):dlm_do_master_request:1334
>>>> ERROR: link to 1 went down!
>>>> Jan 27 14:59:34 nvr2-rc kernel: (6056,1):dlm_do_master_request:1334
>>>> ERROR: link to 1 went down!
>>>> Jan 27 14:59:34 nvr2-rc kernel: (6056,1):dlm_get_lock_resource:917 ERROR:
>>>> status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel: (3344,3):dlm_get_lock_resource:917 ERROR:
>>>> status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel: (4000,4):dlm_purge_lockres:206 ERROR:
>>>>  status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel:
>>>> (5214,4):dlm_send_remote_unlock_request:359 ERROR: status = -112
>>>> Jan 27 14:59:34 nvr2-rc kernel: o2net: accepted connection from node
>>>> nvr1- rc.minint.it (num 1) at 1.1.1.5:7777
>>>> Jan 27 15:29:30 nvr2-rc kernel: o2net: connection to node
>>>> nvr1-rc.minint.it (num 1) at 1.1.1.5:7777 has been idle for 35.0 seconds,
>>>> shutting it down. Jan 27 15:29:30 nvr2-rc kernel:
>>>> (0,5):o2net_idle_timer:1503 here are some times that might help debug the
>>>> situation: (tmr 1264602535.751483 now 1264602570.751723 dr
>>>> 1264602535.751469 adv
>>>>  1264602535.751491:1264602535.751491 func (75554989:505)
>>>>  1264602517.753992:1264602517.753999)
>>>> Jan 27 15:29:30 nvr2-rc kernel: o2net: no longer connected to node nvr1-
>>>> rc.minint.it (num 1) at 1.1.1.5:7777
>>>> Jan 27 15:30:05 nvr2-rc kernel: (3943,1):o2net_connect_expired:1664
>>>> ERROR: no connection established with node 1 after 35.0 seconds, giving
>>>> up and returning errors.
>>>> Jan 27 15:30:40 nvr2-rc kernel: (3943,5):o2net_connect_expired:1664
>>>> ERROR: no connection established with node 1 after 35.0 seconds, giving
>>>> up and returning errors.
>>>> Jan 27 15:30:58 nvr2-rc kernel: (3985,1):ocfs2_dlm_eviction_cb:98 device
>>>> (120,1): dlm has evicted node 1
>>>> Jan 27 15:30:58 nvr2-rc kernel: (6983,5):dlm_get_lock_resource:844
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042:M00000000000000000000122d757fa5: at
>>>> least one node (1) to recover before lock mastery can begin
>>>> Jan 27 15:30:58 nvr2-rc kernel: (6056,1):dlm_restart_lock_mastery:1223
>>>>  ERROR: node down! 1
>>>> Jan 27 15:30:58 nvr2-rc kernel: (6056,1):dlm_wait_for_lock_mastery:1040
>>>>  ERROR: status = -11
>>>> Jan 27 15:30:58 nvr2-rc kernel: (3344,3):dlm_restart_lock_mastery:1223
>>>>  ERROR: node down! 1
>>>> Jan 27 15:30:58 nvr2-rc kernel: (3344,3):dlm_wait_for_lock_mastery:1040
>>>>  ERROR: status = -11
>>>> Jan 27 15:30:59 nvr2-rc kernel: (6983,5):dlm_get_lock_resource:898
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042:M00000000000000000000122d757fa5: at
>>>> least one node (1) to recover before lock mastery can begin
>>>> Jan 27 15:30:59 nvr2-rc kernel: (6056,1):dlm_get_lock_resource:898
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042:N00000000000201de: at least one node (1)
>>>>  to recover before lock mastery can begin
>>>> Jan 27 15:30:59 nvr2-rc kernel: (3344,3):dlm_get_lock_resource:898
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042:M000000000000004dc281b900000000: at
>>>> least one node (1) to recover before lock mastery can begin
>>>> Jan 27 15:31:01 nvr2-rc kernel: (4001,6):dlm_get_lock_resource:844
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042:$RECOVERY: at least one node (1) to
>>>>  recover before lock mastery can begin
>>>> Jan 27 15:31:01 nvr2-rc kernel: (4001,6):dlm_get_lock_resource:878
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042: recovery map is not empty, but must
>>>>  master $RECOVERY lock now
>>>> Jan 27 15:31:01 nvr2-rc kernel: (4001,6):dlm_do_recovery:524 (4001) Node
>>>> 0 is the Recovery Master for the Dead Node 1 for Domain
>>>> 3AE0B7F3BAB749D09D37DAE16FA38042
>>>> Jan 27 15:31:10 nvr2-rc kernel: (6983,1):ocfs2_replay_journal:1183
>>>>  Recovering node 1 from slot 0 on device (120,1)
>>>>
>>>> thanks
>>>> Nicola
>>>>
>>>> In data giovedì 14 gennaio 2010 21:13:15, Sunil Mushran ha scritto:
>>>> : > Mailing List SVR wrote:
>>>> : >
>>>>         
>>>>>> Hi,
>>>>>>
>>>>>> periodically one of on my two nodes cluster is fenced here are the
>>>>>> logs:
>>>>>>
>>>>>> Jan 14 07:01:44 nvr1-rc kernel: o2net: no longer connected to node
>>>>>> nvr2- rc.minint.it (num 0) at 1.1.1.6:7777
>>>>>> Jan 14 07:01:44 nvr1-rc kernel: (21534,1):dlm_do_master_request:1334
>>>>>> ERROR: link to 0 went down!
>>>>>> Jan 14 07:01:44 nvr1-rc kernel: (4007,4):dlm_send_proxy_ast_msg:458
>>>>>> ERROR: status = -112
>>>>>> Jan 14 07:01:44 nvr1-rc kernel: (4007,4):dlm_flush_asts:600 ERROR:
>>>>>> status = -112
>>>>>> Jan 14 07:01:44 nvr1-rc kernel: (21534,1):dlm_get_lock_resource:917
>>>>>> ERROR: status = -112
>>>>>> Jan 14 07:02:19 nvr1-rc kernel: (3950,5):o2net_connect_expired:1664
>>>>>> ERROR: no connection established with node 0 after 35.0 seconds,
>>>>>> giving up and returning errors.
>>>>>> Jan 14 07:02:54 nvr1-rc kernel: (3950,5):o2net_connect_expired:1664
>>>>>> ERROR: no connection established with node 0 after 35.0 seconds,
>>>>>> giving up and returning errors.
>>>>>> Jan 14 07:03:10 nvr1-rc kernel: (4007,4):dlm_send_proxy_ast_msg:458
>>>>>> ERROR: status = -107
>>>>>> Jan 14 07:03:10 nvr1-rc kernel: (4007,4):dlm_flush_asts:600 ERROR:
>>>>>> status = -107
>>>>>> Jan 14 07:03:29 nvr1-rc kernel: (3950,5):o2net_connect_expired:1664
>>>>>> ERROR: no connection established with node 0 after 35.0 seconds,
>>>>>> giving up and returning errors.
>>>>>> Jan 14 07:03:50 nvr1-rc kernel: (31,5):o2quo_make_decision:146 ERROR:
>>>>>> fencing this node because it is connected to a half-quorum of 1 out
>>>>>> of 2 nodes which doesn't include the lowest active node 0
>>>>>> Jan 14 07:03:50 nvr1-rc kernel: (31,5):o2hb_stop_all_regions:1967
>>>>>> ERROR: stopping heartbeat on all active regions.
>>>>>>
>>>>>> I'm sure there are no network connectivity problem but it is possible
>>>>>> that there are heavy IO loads, is this the intended behaviour? Why
>>>>>> under heavy load the loaded node is fenced?
>>>>>>
>>>>>> I'm using ocfs2-1.4.4 on rhel5 kernel-2.6.18-164.6.1.el5
>>>>>>             
>>>>> So the network connection snapped. What it means is that the nodes
>>>>> could not ping each other for 35 seconds. In fact node 1 (this one),
>>>>> tried to reconnect to node 0 but got no reply back. So the network
>>>>> issue lasted for over 2 mins.
>>>>>
>>>>> Switch could be one culprit. See if the switch logs say something.
>>>>> Other possibility is that node 0 was paging heavily. Or kswapd was
>>>>> pegged at 100%. This is hard to determine after the fact. Something to
>>>>> keep in mind the next time you see the same issue. If that is the case,
>>>>> then that needs to be fixed. Maybe add more memory. Or, if you are
>>>>> running the database, ensure you are using hugepages. etc.
>>>>>           
>>> _______________________________________________
>>> Ocfs2-users mailing list
>>> Ocfs2-users at oss.oracle.com
>>> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>>>
>>>       
>> _______________________________________________
>> Ocfs2-users mailing list
>> Ocfs2-users at oss.oracle.com
>> http://oss.oracle.com/mailman/listinfo/ocfs2-users
>>
>>

More information about the Ocfs2-users mailing list