PG: ignore non MISSING pg query in ReplicaActive

1) Replica sends notify
2) Prior to processing notify, primary queues query to replica
3) Primary processes notify and activates sending MOSDPGLog
to replica.
4) Primary does do_notifies at end of process_peering_events
and sends to Query.
5) Replica sees MOSDPGLog and activates
6) Replica sees Query and asserts.

In the above case, the Replica should simply ignore the old
Query.

Fixes: #4050
Backport: bobtail
Signed-off-by: Samuel Just <sam.just@inktank.com>
Reviewed-by: Sage Weil <sage@inktank.com>
(cherry picked from commit 8222cbc8f35c359a35f8381ad90ff0eed5615dac)

diff --git a/src/osd/PG.cc b/src/osd/PG.cc
index 23f1623d7fea4f836101e5d13ded70c0e882bb94..982ce1a3d76751f93ede8c29078f88d4c4b560f4 100644 (file)
--- a/src/osd/PG.cc
+++ b/src/osd/PG.cc
@@ -6239,9 +6239,10 @@ boost::statechart::result PG::RecoveryState::ReplicaActive::react(const ActMap&)
 boost::statechart::result PG::RecoveryState::ReplicaActive::react(const MQuery& query)
 {
   PG *pg = context< RecoveryMachine >().pg;
-  assert(query.query.type == pg_query_t::MISSING);
-  pg->update_history_from_master(query.query.history);
-  pg->fulfill_log(query.from, query.query, query.query_epoch);
+  if (query.query.type == pg_query_t::MISSING) {
+    pg->update_history_from_master(query.query.history);
+    pg->fulfill_log(query.from, query.query, query.query_epoch);
+  } // else: from prior to activation, safe to ignore
   return discard_event();
 }