My evivronment is release 3.1.0, sometimes the nfs service got crash, and its log as following: [2010-12-17 17:51:16.606707] E [afr-self-heal-common.c:1214:sh_missing_entries_create] vmpool-replicate-1: no mis sing files - /TV900000600005/TV900000600005-flat.vmdk. proceeding to metadata check [2010-12-17 17:51:16.608413] E [afr-common.c:110:afr_set_split_brain] vmpool-replicate-1: invalid argument: inode [2010-12-17 17:51:16.608444] I [afr-self-heal-common.c:1526:afr_self_heal_completion_cbk] vmpool-replicate-1: bac kground data self-heal completed on /TV900000600005/TV900000600005-flat.vmdk pending frames: patchset: v3.1.0 signal received: 11 time of crash: 2010-12-17 18:16:32 configuration details: argp 1 backtrace 1 dlfcn 1 fdatasync 1 libpthread 1 llistxattr 1 setfsid 1 spinlock 1 epoll.h 1 xattr.h 1 st_atim.tv_nsec 1 package-string: glusterfs 3.1.0 /lib64/libc.so.6[0x3d924302d0] /lib64/libpthread.so.0(pthread_spin_lock+0x2)[0x3d9300b722] /usr/local/lib/libglusterfs.so.0(fd_unref+0x3b)[0x2ada65d9f8cb] /usr/local/lib/glusterfs/3.1.0/xlator/cluster/replicate.so(afr_local_cleanup+0x56)[0x2aaaaad15136] /usr/local/lib/glusterfs/3.1.0/xlator/cluster/replicate.so(afr_fstat_cbk+0xdc)[0x2aaaaacef5dc] /usr/local/lib/glusterfs/3.1.0/xlator/protocol/client.so(client3_1_fstat_cbk+0x287)[0x2aaaaaacd2c7] /usr/local/lib/libgfrpc.so.0(rpc_clnt_handle_reply+0xa2)[0x2ada65fdf2f2] /usr/local/lib/libgfrpc.so.0(rpc_clnt_notify+0x8d)[0x2ada65fdf4ed] /usr/local/lib/libgfrpc.so.0(rpc_transport_notify+0x2c)[0x2ada65fda78c] /usr/local/lib/glusterfs/3.1.0/rpc-transport/socket.so(socket_event_poll_in+0x3f)[0x2aaabb0ec34f] /usr/local/lib/glusterfs/3.1.0/rpc-transport/socket.so(socket_event_handler+0x168)[0x2aaabb0ec4d8] /usr/local/lib/libglusterfs.so.0[0x2ada65da0ee7] /usr/local/sbin/glusterfs(main+0x37d)[0x4046fd] /lib64/libc.so.6(__libc_start_main+0xf4)[0x3d9241d994] /usr/local/sbin/glusterfs[0x402dd9]
Hi Gluster Fan, please upgrade to 3.1.2 release. It may have been fixed as part of fixes for similar crashes. Thanks.
Resolving, please re-open if the problem still occurs. It hasnt been seen in our tests with self-heal.