glusterd: Fixed glusterd crash in volume add-brick

PROBLEMS:

a. glusterd crashes when add-brick operation fails on the machine other
   than the originator, owing to double free done on op_errstr:
   once in glusterd_op_begin_synctask and once through a dict unref on
   req_dict in gd_sync_task_begin.

b. In gd_sync_task_begin, there's no need to place the error string
   in the dictionary, when it is never retrieved and used elsewhere.

c. Command execution status is not logged into .cmd_log_history.

FIX:

For (a) and (b): Knocked off code that places the error string in req_dict.
That way, both the problems are solved.

For (b), passed op_ctx to glusterd_op_send_cli_response as it is needed to
extract the command string before logging.

Change-Id: I549a07ba5e31332b691a8cacd1ab32c2673810ba
BUG: 862834
Signed-off-by: Krutika Dhananjay <kdhananj@redhat.com>
Reviewed-on: http://review.gluster.org/4102
Reviewed-by: Krishnan Parthasarathi <kparthas@redhat.com>
Reviewed-by: Pranith Kumar Karampuri <pkarampu@redhat.com>
Tested-by: Gluster Build System <jenkins@build.gluster.com>

0 files changed, 0 insertions, 0 deletions