[Gluster-devel] 3.1.3 crash on volume start

[Mark "Naoki" Rogers]

Ok solved it; stems from the dead machine I mentioned. Seems even though 
the machine was dead gluster decided it was "Connected" and all 
subsequent attempts to talk to it (when performing vol or peer commands) 
resulting in a segfault.

I stopped gsluter, deleted the etc/glusterd/peers/ID file from all 
servers and restarted. This gave me back a working setup.

That's a bit ugly though.


On 04/12/2011 04:54 PM, Mark "Naoki" Rogers wrote:
> Attempting to stop an otherwise ok volume also results in a crash;
>
> patchset: v3.1.4qa3
> signal received: 11
> time of crash: 2011-04-12 16:53:46
> 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 3.1.4qa3
> /lib64/libc.so.6(+0x33140)[0x7f7d5dc8e140]
> /opt/glusterfs/3.1.4qa3/lib64/glusterfs/3.1.4qa3/xlator/mgmt/glusterd.so(+0x3323b)[0x7f7d5c4e423b]
> /opt/glusterfs/3.1.4qa3/lib64/glusterfs/3.1.4qa3/xlator/mgmt/glusterd.so(glusterd_op_sm+0xee)[0x7f7d5c4e36ce]
> /opt/glusterfs/3.1.4qa3/lib64/glusterfs/3.1.4qa3/xlator/mgmt/glusterd.so(glusterd_handle_cli_stop_volume+0x96)[0x7f7d5c4d2616]
> /opt/glusterfs/3.1.4qa3/lib64/libgfrpc.so.0(rpcsvc_handle_rpc_call+0x28f)[0x7f7d5e63401f]
> /opt/glusterfs/3.1.4qa3/lib64/libgfrpc.so.0(rpcsvc_notify+0x123)[0x7f7d5e634653]
> /opt/glusterfs/3.1.4qa3/lib64/libgfrpc.so.0(rpc_transport_notify+0x2d)[0x7f7d5e63821d]
> /opt/glusterfs/3.1.4qa3/lib64/glusterfs/3.1.4qa3/rpc-transport/socket.so(socket_event_poll_in+0x34)[0x7f7d5c2a9b44]
> /opt/glusterfs/3.1.4qa3/lib64/glusterfs/3.1.4qa3/rpc-transport/socket.so(socket_event_handler+0xb3)[0x7f7d5c2a9de3]
> /opt/glusterfs/3.1.4qa3/lib64/libglusterfs.so.0(+0x37d42)[0x7f7d5e878d42]
> /opt/glusterfs/3.1.4qa3/sbin/glusterd(main+0x263)[0x4054b3]
> /lib64/libc.so.6(__libc_start_main+0xfd)[0x7f7d5dc79e5d]
> /opt/glusterfs/3.1.4qa3/sbin/glusterd[0x402db9]