diff --git a/.github/workflows/ci.yml b/.github/workflows/ci.yml
index 25fe6f9db..3cd962adb 100644
--- a/.github/workflows/ci.yml
+++ b/.github/workflows/ci.yml
@@ -78,7 +78,7 @@ jobs:
           echo Run ctest -V -L DFLY
           #GLOG_logtostderr=1 GLOG_vmodule=transaction=1,engine_shard_set=1
           GLOG_logtostderr=1 GLOG_vmodule=rdb_load=1,rdb_save=1,snapshot=1 ctest -V -L DFLY
-          ./dragonfly_test  --mem_defrag_threshold=0.05 # trying to catch issue with defrag
+          ./dragonfly_test  --gtest_repeat=100
           # GLOG_logtostderr=1 GLOG_vmodule=transaction=1,engine_shard_set=1 CTEST_OUTPUT_ON_FAILURE=1 ninja server/test
   lint-test-chart:
     runs-on: ubuntu-latest
diff --git a/src/server/dragonfly_test.cc b/src/server/dragonfly_test.cc
index b920646ba..a2e42a636 100644
--- a/src/server/dragonfly_test.cc
+++ b/src/server/dragonfly_test.cc
@@ -471,6 +471,33 @@ TEST_F(DflyEngineTest, EvalBug713) {
   fb0.Join();
 }
 
+// Tests deadlock that happenned due to a fact that trans->Schedule was called
+// before interpreter->Lock().
+//
+// The problematic scenario:
+// 1. transaction 1 schedules itself and blocks on an interpreter lock
+// 2. transaction 2 schedules itself, but meanwhile an interpreter unlocks itself and
+//    transaction 2 grabs the lock but can not progress due to transaction 1 already
+//    scheduled before.
+TEST_F(DflyEngineTest, EvalBug713b) {
+  const char* script = "return redis.call('get', KEYS[1])";
+
+  const uint32_t kNumFibers = 20;
+  fibers_ext::Fiber fibers[kNumFibers];
+
+  for (unsigned j = 0; j < kNumFibers; ++j) {
+    fibers[j] = pp_->at(1)->LaunchFiber([=, this] {
+      for (unsigned i = 0; i < 50; ++i) {
+        Run(StrCat("fb", j), {"eval", script, "3", kKeySid0, kKeySid1, kKeySid2});
+      }
+    });
+  }
+
+  for (unsigned j = 0; j < kNumFibers; ++j) {
+    fibers[j].Join();
+  }
+}
+
 TEST_F(DflyEngineTest, EvalSha) {
   auto resp = Run({"script", "load", "return 5"});
   EXPECT_THAT(resp, ArgType(RespExpr::STRING));
diff --git a/src/server/main_service.cc b/src/server/main_service.cc
index b5e9d5508..d74595573 100644
--- a/src/server/main_service.cc
+++ b/src/server/main_service.cc
@@ -906,6 +906,8 @@ void Service::Unwatch(CmdArgList args, ConnectionContext* cntx) {
 
 void Service::CallFromScript(CmdArgList args, ObjectExplorer* reply, ConnectionContext* cntx) {
   DCHECK(cntx->transaction);
+  DVLOG(1) << "CallFromScript " << cntx->transaction->DebugId() << " " << ArgS(args, 0);
+
   InterpreterReplier replier(reply);
   facade::SinkReplyBuilder* orig = cntx->Inject(&replier);
 
@@ -1013,11 +1015,11 @@ void Service::EvalInternal(const EvalArgs& eval_args, Interpreter* interpreter,
   }
   DCHECK(cntx->transaction);
 
+  auto lk = interpreter->Lock();
+
   if (!eval_args.keys.empty())
     cntx->transaction->Schedule();
 
-  auto lk = interpreter->Lock();
-
   interpreter->SetGlobalArray("KEYS", eval_args.keys);
   interpreter->SetGlobalArray("ARGV", eval_args.args);
   interpreter->SetRedisFunc(
@@ -1039,12 +1041,12 @@ void Service::EvalInternal(const EvalArgs& eval_args, Interpreter* interpreter,
   CHECK(result == Interpreter::RUN_OK);
 
   EvalSerializer ser{static_cast<RedisReplyBuilder*>(cntx->reply_builder())};
-
   if (!interpreter->IsResultSafe()) {
     (*cntx)->SendError("reached lua stack limit");
   } else {
     interpreter->SerializeResult(&ser);
   }
+
   interpreter->ResetStack();
 }
 
diff --git a/src/server/transaction.cc b/src/server/transaction.cc
index 5498cce2c..88fe86f6b 100644
--- a/src/server/transaction.cc
+++ b/src/server/transaction.cc
@@ -904,7 +904,7 @@ pair<bool, bool> Transaction::ScheduleInShard(EngineShard* shard) {
     // All transactions in the queue must acquire the intent lock.
     lock_granted = shard->db_slice().Acquire(mode, lock_args) && shard_unlocked;
     sd.local_mask |= KEYLOCK_ACQUIRED;
-    DVLOG(1) << "Lock granted " << lock_granted << " for trans " << DebugId();
+    DVLOG(2) << "Lock granted " << lock_granted << " for trans " << DebugId();
   }
 
   if (!txq->Empty()) {