3 files changed, 26 insertions, 6 deletions

diff --git a/fs/ocfs2/dlm/dlmdomain.c b/fs/ocfs2/dlm/dlmdomain.c
index 02d315fef432..50a59d2337b2 100644
--- a/fs/ocfs2/dlm/dlmdomain.c
+++ b/fs/ocfs2/dlm/dlmdomain.c
@@ -877,7 +877,7 @@ static int dlm_query_join_handler(struct o2net_msg *msg, u32 len, void *data,
 	 * to be put in someone's domain map.
 	 * Also, explicitly disallow joining at certain troublesome
 	 * times (ie. during recovery). */
-	if (dlm && dlm->dlm_state != DLM_CTXT_LEAVING) {
+	if (dlm->dlm_state != DLM_CTXT_LEAVING) {
 		int bit = query->node_idx;
 		spin_lock(&dlm->spinlock);
 
diff --git a/fs/ocfs2/dlm/dlmmaster.c b/fs/ocfs2/dlm/dlmmaster.c
index 215e41abf101..3689b3592042 100644
--- a/fs/ocfs2/dlm/dlmmaster.c
+++ b/fs/ocfs2/dlm/dlmmaster.c
@@ -1460,6 +1460,18 @@ way_up_top:
 
 		/* take care of the easy cases up front */
 		spin_lock(&res->spinlock);
+
+		/*
+		 * Right after dlm spinlock was released, dlm_thread could have
+		 * purged the lockres. Check if lockres got unhashed. If so
+		 * start over.
+		 */
+		if (hlist_unhashed(&res->hash_node)) {
+			spin_unlock(&res->spinlock);
+			dlm_lockres_put(res);
+			goto way_up_top;
+		}
+
 		if (res->state & (DLM_LOCK_RES_RECOVERING|
 				  DLM_LOCK_RES_MIGRATING)) {
 			spin_unlock(&res->spinlock);
diff --git a/fs/ocfs2/dlm/dlmrecovery.c b/fs/ocfs2/dlm/dlmrecovery.c
index 3365839d2971..79b5af5e6a7b 100644
--- a/fs/ocfs2/dlm/dlmrecovery.c
+++ b/fs/ocfs2/dlm/dlmrecovery.c
@@ -1656,14 +1656,18 @@ int dlm_do_master_requery(struct dlm_ctxt *dlm, struct dlm_lock_resource *res,
 	req.namelen = res->lockname.len;
 	memcpy(req.name, res->lockname.name, res->lockname.len);
 
+resend:
 	ret = o2net_send_message(DLM_MASTER_REQUERY_MSG, dlm->key,
 				 &req, sizeof(req), nodenum, &status);
-	/* XXX: negative status not handled properly here. */
 	if (ret < 0)
 		mlog(ML_ERROR, "Error %d when sending message %u (key "
 		     "0x%x) to node %u\n", ret, DLM_MASTER_REQUERY_MSG,
 		     dlm->key, nodenum);
-	else {
+	else if (status == -ENOMEM) {
+		mlog_errno(status);
+		msleep(50);
+		goto resend;
+	} else {
 		BUG_ON(status < 0);
 		BUG_ON(status > DLM_LOCK_RES_OWNER_UNKNOWN);
 		*real_master = (u8) (status & 0xff);
@@ -1705,9 +1709,13 @@ int dlm_master_requery_handler(struct o2net_msg *msg, u32 len, void *data,
 			int ret = dlm_dispatch_assert_master(dlm, res,
 							     0, 0, flags);
 			if (ret < 0) {
-				mlog_errno(-ENOMEM);
-				/* retry!? */
-				BUG();
+				mlog_errno(ret);
+				spin_unlock(&res->spinlock);
+				dlm_lockres_put(res);
+				spin_unlock(&dlm->spinlock);
+				dlm_put(dlm);
+				/* sender will take care of this and retry */
+				return ret;
 			} else
 				__dlm_lockres_grab_inflight_worker(dlm, res);
 			spin_unlock(&res->spinlock);