@@ -787,6 +787,7 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from,
{
struct dlm_lock_request lr;
int ret;
+ int status;
mlog(0, "\n");
@@ -800,13 +801,15 @@ static int dlm_request_all_locks(struct dlm_ctxt *dlm, u8 request_from,
// send message
ret = o2net_send_message(DLM_LOCK_REQUEST_MSG, dlm->key,
- &lr, sizeof(lr), request_from, NULL);
+ &lr, sizeof(lr), request_from, &status);
/* negative status is handled by caller */
if (ret < 0)
mlog(ML_ERROR, "%s: Error %d send LOCK_REQUEST to node %u "
"to recover dead node %u\n", dlm->name, ret,
request_from, dead_node);
+ else
+ ret = status;
// return from here, then
// sleep until all received or error
return ret;
dlm_request_all_locks() should deal with the status sent from target node if DLM_LOCK_REQUEST_MSG is sent successfully, or recovery master will fall into endless loop, waiting for other nodes to send locks and DLM_RECO_DATA_DONE_MSG to me. NodeA NodeB selected as recovery master dlm_remaster_locks() ->dlm_request_all_locks() send DLM_LOCK_REQUEST_MSG to nodeA It happened that NodeA cannot alloc memory when it processes this message. dlm_request_all_locks_handler() do not queue dlm_request_all_locks_worker and returns -ENOMEM. It will never send locks andDLM_RECO_DATA_DONE_MSG to NodeB. NodeB do not deal with the status sent from nodeA, and will fall in endless loop waiting for the recovery state of NodeA to be changed. Signed-off-by: joyce <xuejiufei@huawei.com> --- fs/ocfs2/dlm/dlmrecovery.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)