[Gluster-users] glusterfs crash when the one of replicate node restart
krish
kp at gluster.com
Mon Jan 2 11:48:45 UTC 2012
Hi Changliang,
Could you attach the logs of the servers (bricks)?
>
>> Because to keep availability,we haven't strace the
>> process.After shudowning the damon,the cluster recover.
>> In our case,
>
Pranith was asking for a core dump file or the backtrace (of the crashed
process), not strace output.
thanks,
krish
>
>> 10.1.1.64(dfs-client-6): online node,when the other node(65)
>> restart,cpu usr usage reach 100% (glusterfsd process)
>> 10.1.1.65(dfs-client-7): offline node,when it restart,the
>> client nfs mount point unavailable.
>> The nfs.log show that the reason of issue will be cause
>> by client-6 high cpu usage,there are lots of error like:
>>
>> [2011-12-14 13:25:53.30308] E [rpc-clnt.c:197:call_bail]
>> 0-19loudfs-client-6: bailing out frame type(GlusterFS 3.1)
>> op(XATTROP(33)) xid = 0x89279937x sent = 2011-12-14 13:25:20.
>> 346007. timeout = 30
>>
>>
>>
>>
>>
>>
>>
>> On Wed, Dec 14, 2011 at 6:49 PM, Pranith Kumar K
>> <pranithk at gluster.com <mailto:pranithk at gluster.com>> wrote:
>>
>> On 12/14/2011 03:06 PM, Changliang Chen wrote:
>>> Hi,we have use glusterfs for two years. After upgraded to
>>> 3.2.5,we discover that when one of replicate node reboot and
>>> startup the glusterd daemon,the gluster will crash cause by
>>> the other
>>>
>>> replicate node cpu usage reach 100%.
>>>
>>> Our gluster info:
>>>
>>> Type: Distributed-Replicate
>>> Status: Started
>>> Number of Bricks: 5 x 2 = 10
>>> Transport-type: tcp
>>> Options Reconfigured:
>>> performance.cache-size: 3GB
>>> performance.cache-max-file-size: 512KB
>>> network.frame-timeout: 30
>>> network.ping-timeout: 25
>>> cluster.min-free-disk: 10%
>>>
>>> Our device：
>>>
>>> Dell R710
>>> 600Gsas *6
>>> 3*8Gmem
>>>
>>> The error info:
>>>
>>> [2011-12-14 13:24:10.483812] E [rdma.c:4813:init]
>>> 0-rdma.management: Failed to initialize IB Device
>>> [2011-12-14 13:24:10.483828] E
>>> [rpc-transport.c:742:rpc_transport_load] 0-rpc-transport:
>>> 'rdma' initialization failed
>>> [2011-12-14 13:24:10.483841] W
>>> [rpcsvc.c:1288:rpcsvc_transport_create] 0-rpc-service:
>>> cannot create listener, initing the transport failed
>>> [2011-12-14 13:24:11.967621] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-0
>>> [2011-12-14 13:24:11.967665] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-1
>>> [2011-12-14 13:24:11.967681] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-2
>>> [2011-12-14 13:24:11.967695] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-3
>>> [2011-12-14 13:24:11.967709] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-4
>>> [2011-12-14 13:24:11.967723] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-5
>>> [2011-12-14 13:24:11.967736] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-6
>>> [2011-12-14 13:24:11.967750] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-7
>>> [2011-12-14 13:24:11.967764] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-8
>>> [2011-12-14 13:24:11.967777] E
>>> [glusterd-store.c:1820:glusterd_store_retrieve_volume] 0-:
>>> Unknown key: brick-9
>>> [2011-12-14 13:24:12.465565] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.17:1013
>>> <http://10.1.1.17:1013>)
>>> [2011-12-14 13:24:12.465623] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.8:1013
>>> <http://10.1.1.8:1013>)
>>> [2011-12-14 13:24:12.465656] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.10:1013
>>> <http://10.1.1.10:1013>)
>>> [2011-12-14 13:24:12.465686] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.11:1013
>>> <http://10.1.1.11:1013>)
>>> [2011-12-14 13:24:12.465716] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.125:1013
>>> <http://10.1.1.125:1013>)
>>> [2011-12-14 13:24:12.633288] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.65:1006
>>> <http://10.1.1.65:1006>)
>>> [2011-12-14 13:24:13.138150] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.1:1013
>>> <http://10.1.1.1:1013>)
>>> [2011-12-14 13:24:13.284665] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.3:1013
>>> <http://10.1.1.3:1013>)
>>> [2011-12-14 13:24:15.790805] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.8:1013
>>> <http://10.1.1.8:1013>)
>>> [2011-12-14 13:24:16.113430] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.125:1013
>>> <http://10.1.1.125:1013>)
>>> [2011-12-14 13:24:16.259040] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.10:1013
>>> <http://10.1.1.10:1013>)
>>> [2011-12-14 13:24:16.392058] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.17:1013
>>> <http://10.1.1.17:1013>)
>>> [2011-12-14 13:24:16.429444] W
>>> [socket.c:1494:__socket_proto_state_machine]
>>> 0-socket.management: reading from socket failed. Error
>>> (Transport endpoint is not connected), peer (10.1.1.11:1013
>>> <http://10.1.1.11:1013>)
>>> [2011-12-14 13:26:05.787680] W
>>> [glusterfsd.c:727:cleanup_and_exit]
>>> (-->/lib64/libc.so.6(clone+0x6d) [0x37c8ed3c2d]
>>> (-->/lib64/libpthread.so.0 [0x37c96064a7]
>>> (-->/opt/glusterfs/3.2.5/sbin/glusterd(glusterfs_sigwaiter+0x17c)
>>> [0x40477c]))) 0-: received signum (15), shutting down
>>>
>>> --
>>>
>>> Regards,
>>>
>>> Cocl
>>>
>>>
>>>
>>> _______________________________________________
>>> Gluster-users mailing list
>>> Gluster-users at gluster.org <mailto:Gluster-users at gluster.org>
>>> http://gluster.org/cgi-bin/mailman/listinfo/gluster-users
>> hi Changliang,
>> Could you specify which process crashed. Is it
>> glusterd or glusterfs? Could you provide the stack trace that
>> is present in it's respective logfile. I dont see any stack
>> trace in the logs you have provided.
>>
>> Pranith
>>
>>
>>
>>
>> --
>>
>> Regards,
>>
>> Cocl
>> OM manager
>> 19lou Operation & Maintenance Dept
> Could you send the logs of all the machines, we will check and
> getback to you.
>
> Pranith
>
>
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://supercolony.gluster.org/pipermail/gluster-users/attachments/20120102/176e28b6/attachment.html>
More information about the Gluster-users
mailing list