[Gluster-devel] gluster client crash

Karl Bernard karl at vomba.com
Mon Dec 17 12:36:17 UTC 2007


Here's the backtrace:

First Crash:

Program terminated with signal 11, Segmentation fault.
#0  0x00000000 in ?? ()
(gdb) bt
#0  0x00000000 in ?? ()
#1  0x0032cc0c in ra_close_cbk (frame=0x83c97e4, cookie=0xb4b0c9f0, 
this=0x0, op_ret=0, op_errno=0) at read-ahead.c:256
#2  0x00196c1d in wb_ffr_cbk (frame=0xb4b0c9f0, cookie=0xb4b0ca28, 
this=0x838e6d0, op_ret=0, op_errno=0) at write-behind.c:693
#3  0x00262dab in iot_close_cbk (frame=0xb4b0ca28, cookie=0xb4ea8788, 
this=0x838e688, op_ret=0, op_errno=0) at io-threads.c:174
#4  0x00118b46 in afr_close_cbk (frame=0xb4ea8788, cookie=0x9c5c470, 
this=0x838e2c0, op_ret=0, op_errno=0) at afr.c:3221
#5  0x0013367d in client_close_cbk (frame=0x9c5c470, args=0x9c5ba60) at 
client-protocol.c:3436
#6  0x001374c4 in notify (this=0x838dbb8, event=2, data=0x83c7c68) at 
client-protocol.c:4568
#7  0x006c9717 in transport_notify (this=0x83c97e4, event=138188772) at 
transport.c:154
#8  0x006ca473 in sys_epoll_iteration (ctx=0xbfebc014) at epoll.c:54
#9  0x006c984c in poll_iteration (ctx=0xbfebc014) at transport.c:302
#10 0x0804a494 in main (argc=8, argv=0xbfebc114) at glusterfs.c:400

 From the log:
---------
got signal (11), printing backtrace
---------
[0xcc5420]
/usr/local/lib/glusterfs/1.3.8/xlator/performance/write- behind.so[0x196c1d]
/usr/local/lib/glusterfs/1.3.8/xlator/performance/io-threads.so[0x262dab]
/usr/local/lib/glusterfs/1.3.8/xlator/cluster/afr.so(afr_close_cbk+0x1d6)[0x118b46]
/usr/local/lib/glusterfs/1.3.8/xlator/protocol/client.so[0x13367d]
/usr/local/lib/glusterfs/1.3.8/xlator/protocol/client.so(notify+0xa84)[0x1374c4]
/usr/local/lib/libglusterfs.so.0(transport_notify+0x37)[0x6c9717]
/usr/local/lib/libglusterfs.so.0(sys_epoll_iteration+0xf3)[0x6ca473]
/usr/local/lib/libglusterfs.so.0(poll_iteration+0x7c)[0x6c984c]
[glusterfs](main+0x424)[0x804a494]
/lib/libc.so.6(__libc_start_main+0xdc)[0xa49dec]
[glusterfs][0x8049fe1]


------------------------------------
Second crash:
#0  0x0000003ff806ea75 in _int_malloc () from /lib64/libc.so.6
#1  0x0000003ff80706cd in malloc () from /lib64/libc.so.6
#2  0x00002aaaaaacbfe2 in gf_block_unserialize_transport 
(trans=0x15f1eed0, max_block_size=268435456) at protocol.c:344
#3  0x00002aaaaaf00544 in notify (this=0x15f1bb30, event=<value 
optimized out>, data=0x15f1eed0) at client-protocol.c:4877
#4  0x00002aaaaaaccdf5 in sys_epoll_iteration (ctx=0x7fff26e3a940) at 
epoll.c:54
#5  0x00002aaaaaacc305 in poll_iteration (ctx=0x7fff26e3a940) at 
transport.c:302
#6  0x000000000040307d in main (argc=8, argv=0x7fff26e3aae8) at 
glusterfs.c:400

 From the log:
---------
got signal (11), printing backtrace
---------
/lib64/libc.so.6[0x3ff8030070]
/lib64/libc.so.6[0x3ff806ea75]
/lib64/libc.so.6(__libc_malloc+0x7d)[0x3ff80706cd]
/usr/local/lib/libglusterfs.so.0(gf_block_unserialize_transport+0x3d2)[0x2aaaaaacbfe2]
/usr/local/lib/glusterfs/1.3.8/xlator/protocol/client.so(notify+0x244)[0x2aaaaaf00544]
/usr/local/lib/libglusterfs.so.0(sys_epoll_iteration+0xd5)[0x2aaaaaaccdf5]
/usr/local/lib/libglusterfs.so.0(poll_iteration+0x75)[0x2aaaaaacc305]
[glusterfs](main+0x38d)[0x40307d]
/lib64/libc.so.6(__libc_start_main+0xf4)[0x3ff801d8a4]
[glusterfs][0x402c59]

>
>
>     Anand Avati wrote:
>>     Karl,
>>      can  you get a backtrace from the coredump with gdb please? that
>>     would help a lot.
>>
>>     avati
>>
>>     2007/12/17, Karl Bernard <karl at vomba.com <mailto:karl at vomba.com> >:
>>
>>
>>         The client crashed, if this can be helpful:
>>
>>         2007-12-15 17:03:59 W [client-protocol.c
>>         :289:client_protocol_xfer]
>>         sxx01: attempting to pipeline request type(0) op(34) with
>>         handshake
>>
>>         ---------
>>         got signal (11), printing backtrace
>>         ---------
>>         [0xcc5420]
>>         /usr/local/lib/glusterfs/1.3.8/xlator/performance/write-
>>         behind.so[0x196c1d]
>>         /usr/local/lib/glusterfs/1.3.8/xlator/performance/io-threads.so[0x262dab]
>>         /usr/local/lib/glusterfs/1.3.8/xlator/cluster/afr.so(afr_close_cbk+0x1d6)[0x118b46]
>>         /usr/local/lib/glusterfs/1.3.8/xlator/protocol/client.so[0x13367d]
>>
>>         /usr/local/lib/glusterfs/1.3.8/xlator/protocol/client.so(notify+0xa84)[0x1374c4]
>>         /usr/local/lib/libglusterfs.so.0(transport_notify+0x37)[0x6c9717]
>>         /usr/local/lib/libglusterfs.so.0(sys_epoll_iteration+0xf3)[0x6ca473]
>>
>>         /usr/local/lib/libglusterfs.so.0(poll_iteration+0x7c)[0x6c984c]
>>         [glusterfs](main+0x424)[0x804a494]
>>         /lib/libc.so.6(__libc_start_main+0xdc)[0xa49dec]
>>         [glusterfs][0x8049fe1]
>>
>>
>>         glusterfs 1.3.8
>>         installed from tla, last patch:
>>         2007-12-03 22:29:15 GMT Anand V. Avati <avati at 80x25.org
>>         <mailto:avati at 80x25.org>>        patch-594
>>
>>         Config client:
>>         ----------------------------------------------------------
>>         volume sxx01
>>         type protocol/client
>>         option transport-type tcp/client
>>         option remote-host sxx01b
>>         option remote-subvolume brick
>>         end-volume
>>
>>         volume sxx02
>>         type protocol/client
>>         option transport-type tcp/client
>>         option remote-host sxx02b
>>         option remote-subvolume brick
>>         end-volume
>>
>>         volume afr1-2
>>           type cluster/afr
>>           subvolumes sxx01 sxx02
>>         end-volume
>>
>>         volume iot
>>         type performance/io-threads
>>         subvolumes afr1-2
>>         option thread-count 8
>>         end-volume
>>
>>         ## Add writebehind feature
>>         volume writebehind
>>           type performance/write-behind
>>           option aggregate-size 128kB
>>           subvolumes iot
>>         end-volume
>>
>>         ## Add readahead feature
>>         volume readahead
>>           type performance/read-ahead
>>           option page-size 256kB     #
>>           option page-count 16       # cache per file  = (page-count
>>         x page-size)
>>           subvolumes writebehind
>>         end-volume
>>
>>         ------------------------------------------------------
>>
>>         Config Server:
>>         volume brick-posix
>>                 type storage/posix
>>                 option directory /data/glusterfs/dataspace
>>         end-volume
>>
>>         volume brick-ns
>>                 type storage/posix
>>                 option directory /data/glusterfs/namespace
>>         end-volume
>>
>>         volume brick
>>           type performance/io-threads
>>           option thread-count 2
>>           option cache-size 32MB
>>           subvolumes brick-posix
>>         end-volume
>>
>>         volume server
>>                 type protocol/server
>>                 option transport-type tcp/server
>>                 subvolumes brick brick-ns
>>                 option auth.ip.brick.allow 172.16.93.*
>>                 option auth.ip.brick-ns.allow 172.16.93.*
>>         end-volume
>>
>>         ------------------------------------
>>
>>         The client was most likely  checking for the existence of a
>>         file or
>>         writing a new file to the servers.
>>
>>
>>
>>
>>     -- 
>>     If I traveled to the end of the rainbow
>>     As Dame Fortune did intend,
>>     Murphy would be there to tell me
>>     The pot's at the other end. 
>
>
>
>
> -- 
> If I traveled to the end of the rainbow
> As Dame Fortune did intend,
> Murphy would be there to tell me
> The pot's at the other end. 




More information about the Gluster-devel mailing list