diff --git a/firedrake/variational_solver.py b/firedrake/variational_solver.py
index 3c9a2689a4..609d599800 100644
--- a/firedrake/variational_solver.py
+++ b/firedrake/variational_solver.py
@@ -300,7 +300,7 @@ def solve(self, bounds=None):
         coefficients = utils.unique(chain.from_iterable(form.coefficients() for form in forms if form is not None))
         # Make sure the solution dm is visited last
         solution_dm = self.snes.getDM()
-        problem_dms = [V.dm for V in utils.unique(c.function_space() for c in coefficients) if V.dm != solution_dm]
+        problem_dms = [V.dm for V in utils.unique(chain.from_iterable(c.function_space() for c in coefficients)) if V.dm != solution_dm]
         problem_dms.append(solution_dm)
 
         for dbc in problem.dirichlet_bcs():
diff --git a/tests/regression/test_star_pc.py b/tests/regression/test_star_pc.py
index b636dc9710..38ad450514 100644
--- a/tests/regression/test_star_pc.py
+++ b/tests/regression/test_star_pc.py
@@ -1,4 +1,5 @@
 import pytest
+import warnings
 from firedrake import *
 from firedrake.petsc import DEFAULT_DIRECT_SOLVER
 try:
@@ -20,6 +21,12 @@ def backend(request):
     return request.param
 
 
+def filter_warnings(caller):
+    with warnings.catch_warnings():
+        warnings.filterwarnings("error", "Creating new TransferManager", RuntimeWarning)
+        caller()
+
+
 def test_star_equivalence(problem_type, backend):
     distribution_parameters = {"partition": True,
                                "overlap_type": (DistributedMeshOverlapType.VERTEX, 1)}
@@ -171,12 +178,12 @@ def test_star_equivalence(problem_type, backend):
     star_params["mg_levels_pc_star_mat_ordering_type"] = "rcm"
     nvproblem = NonlinearVariationalProblem(a, u, bcs=bcs)
     star_solver = NonlinearVariationalSolver(nvproblem, solver_parameters=star_params, nullspace=nsp)
-    star_solver.solve()
+    filter_warnings(star_solver.solve)
     star_its = star_solver.snes.getLinearSolveIterations()
 
     u.assign(0)
     comp_solver = NonlinearVariationalSolver(nvproblem, solver_parameters=comp_params, nullspace=nsp)
-    comp_solver.solve()
+    filter_warnings(comp_solver.solve)
     comp_its = comp_solver.snes.getLinearSolveIterations()
 
     assert star_its == comp_its
@@ -348,12 +355,12 @@ def test_vanka_equivalence(problem_type):
     vanka_params["mg_levels_pc_vanka_mat_ordering_type"] = "rcm"
     nvproblem = NonlinearVariationalProblem(a, u, bcs=bcs)
     star_solver = NonlinearVariationalSolver(nvproblem, solver_parameters=vanka_params, nullspace=nsp)
-    star_solver.solve()
+    filter_warnings(star_solver.solve)
     star_its = star_solver.snes.getLinearSolveIterations()
 
     u.assign(0)
     comp_solver = NonlinearVariationalSolver(nvproblem, solver_parameters=comp_params, nullspace=nsp)
-    comp_solver.solve()
+    filter_warnings(comp_solver.solve)
     comp_its = comp_solver.snes.getLinearSolveIterations()
 
     assert star_its == comp_its