if a node fails to become frozen during recovery, mark it up with as a culprit so...

diff --git a/server/ctdb_recoverd.c b/server/ctdb_recoverd.c
index 2be53f84daef919065518d6e3c572b5d1fb941df..4222703172dfee165ac92a709716d52f9e94d0bc 100644 (file)
--- a/server/ctdb_recoverd.c
+++ b/server/ctdb_recoverd.c
@@ -245,10 +245,18 @@ static int update_capabilities(struct ctdb_context *ctdb, struct ctdb_node_map *
        return 0;
 }
 
+static void set_recmode_fail_callback(struct ctdb_context *ctdb, uint32_t node_pnn, int32_t res, TDB_DATA outdata, void *callback_data)
+{
+       struct ctdb_recoverd *rec = talloc_get_type(callback_data, struct ctdb_recoverd);
+
+       DEBUG(DEBUG_ERR,("Failed to freeze node %u during recovery. Set it as ban culprit for %d credits\n", node_pnn, rec->nodemap->num));
+       ctdb_set_culprit_count(rec, node_pnn, rec->nodemap->num);
+}
+
 /*
   change recovery mode on all nodes
  */
-static int set_recovery_mode(struct ctdb_context *ctdb, struct ctdb_node_map *nodemap, uint32_t rec_mode)
+static int set_recovery_mode(struct ctdb_context *ctdb, struct ctdb_recoverd *rec, struct ctdb_node_map *nodemap, uint32_t rec_mode)
 {
        TDB_DATA data;
        uint32_t *nodes;
@@ -263,8 +271,9 @@ static int set_recovery_mode(struct ctdb_context *ctdb, struct ctdb_node_map *no
                if (ctdb_client_async_control(ctdb, CTDB_CONTROL_FREEZE,
                                                nodes, CONTROL_TIMEOUT(),
                                                false, tdb_null,
-                                               NULL, NULL,
-                                               NULL) != 0) {
+                                               NULL,
+                                               set_recmode_fail_callback,
+                                               rec) != 0) {
                        DEBUG(DEBUG_ERR, (__location__ " Unable to freeze nodes. Recovery failed.\n"));
                        talloc_free(tmp_ctx);
                        return -1;
@@ -1228,7 +1237,7 @@ static int do_recovery(struct ctdb_recoverd *rec,
 
 
        /* set recovery mode to active on all nodes */
-       ret = set_recovery_mode(ctdb, nodemap, CTDB_RECOVERY_ACTIVE);
+       ret = set_recovery_mode(ctdb, rec, nodemap, CTDB_RECOVERY_ACTIVE);
        if (ret != 0) {
                DEBUG(DEBUG_ERR, (__location__ " Unable to set recovery mode to active on cluster\n"));
                return -1;
@@ -1374,7 +1383,7 @@ static int do_recovery(struct ctdb_recoverd *rec,
        DEBUG(DEBUG_NOTICE, (__location__ " Recovery - updated flags\n"));
 
        /* disable recovery mode */
-       ret = set_recovery_mode(ctdb, nodemap, CTDB_RECOVERY_NORMAL);
+       ret = set_recovery_mode(ctdb, rec, nodemap, CTDB_RECOVERY_NORMAL);
        if (ret != 0) {
                DEBUG(DEBUG_ERR, (__location__ " Unable to set recovery mode to normal on cluster\n"));
                return -1;
@@ -1858,7 +1867,7 @@ static void force_election(struct ctdb_recoverd *rec, uint32_t pnn,
        DEBUG(DEBUG_INFO,(__location__ " Force an election\n"));
 
        /* set all nodes to recovery mode to stop all internode traffic */
-       ret = set_recovery_mode(ctdb, nodemap, CTDB_RECOVERY_ACTIVE);
+       ret = set_recovery_mode(ctdb, rec, nodemap, CTDB_RECOVERY_ACTIVE);
        if (ret != 0) {
                DEBUG(DEBUG_ERR, (__location__ " Unable to set recovery mode to active on cluster\n"));
                return;