Reserve memory for other processes on the content node the proper way by including it in hwinfo memory.

This removes the need to scale the write filter limit used for memory.

1 files changed, 5 insertions, 5 deletions

diff --git a/config-model/src/test/java/com/yahoo/config/model/provision/ModelProvisioningTest.java b/config-model/src/test/java/com/yahoo/config/model/provision/ModelProvisioningTest.java
index 37619d3afc6..be272e05bf5 100644
--- a/config-model/src/test/java/com/yahoo/config/model/provision/ModelProvisioningTest.java
+++ b/config-model/src/test/java/com/yahoo/config/model/provision/ModelProvisioningTest.java
@@ -47,6 +47,8 @@ import java.util.stream.Collectors;
 
 import static com.yahoo.config.model.test.TestUtil.joinLines;
 import static com.yahoo.vespa.defaults.Defaults.getDefaults;
+import static com.yahoo.vespa.model.search.NodeResourcesTuning.GB;
+import static com.yahoo.vespa.model.search.NodeResourcesTuning.reservedMemoryGb;
 import static org.hamcrest.CoreMatchers.is;
 import static org.junit.Assert.assertEquals;
 import static org.junit.Assert.assertFalse;
@@ -268,7 +270,7 @@ public class ModelProvisioningTest {
             assertEquals("Heap size is lowered with combined clusters",
                          17, physicalMemoryPercentage(model.getContainerClusters().get("container1")));
             assertEquals("Memory for proton is lowered to account for the jvm heap",
-                         (long)(3 * (Math.pow(1024, 3)) * (1 - 0.17)), protonMemorySize(model.getContentClusters().get("content1")));
+                         (long)((3 - reservedMemoryGb) * (Math.pow(1024, 3)) * (1 - 0.17)), protonMemorySize(model.getContentClusters().get("content1")));
             assertProvisioned(0, ClusterSpec.Id.from("container1"), ClusterSpec.Type.container, model);
             assertProvisioned(2, ClusterSpec.Id.from("content1"), ClusterSpec.Id.from("container1"), ClusterSpec.Type.combined, model);
         }
@@ -304,7 +306,7 @@ public class ModelProvisioningTest {
             assertEquals("Heap size is normal",
                          60, physicalMemoryPercentage(model.getContainerClusters().get("container1")));
             assertEquals("Memory for proton is normal",
-                         (long)(3 * (Math.pow(1024, 3))), protonMemorySize(model.getContentClusters().get("content1")));
+                         (long)((3 - reservedMemoryGb) * (Math.pow(1024, 3))), protonMemorySize(model.getContentClusters().get("content1")));
         }
     }
 
@@ -2000,12 +2002,10 @@ public class ModelProvisioningTest {
         ProtonConfig cfg = getProtonConfig(model, cluster.getSearchNodes().get(0).getConfigId());
         assertEquals(2000, cfg.flush().memory().maxtlssize()); // from config override
         assertEquals(1000, cfg.flush().memory().maxmemory()); // from explicit tuning
-        assertEquals((long) 16 * GB, cfg.flush().memory().each().maxmemory()); // from default node flavor tuning
+        assertEquals((long) (128 - reservedMemoryGb) * GB / 8, cfg.flush().memory().each().maxmemory()); // from default node flavor tuning
         assertEquals(0.92, cfg.writefilter().memorylimit(), 0.0001); // from explicit resource-limits
     }
 
-    private static long GB = 1024 * 1024 * 1024;
-
     private static ProtonConfig getProtonConfig(VespaModel model, String configId) {
         ProtonConfig.Builder builder = new ProtonConfig.Builder();
         model.getConfig(builder, configId);