YARN-4280. CapacityScheduler reservations may not prevent indefinite postponement on a busy cluster. Contributed by Kuhu Shukla
(cherry picked from commit 4d92aefd35
)
This commit is contained in:
parent
a5d12d9c1f
commit
9a0ac56a5c
|
@ -34,36 +34,47 @@ public class CSAssignment {
|
|||
public static final CSAssignment NULL_ASSIGNMENT =
|
||||
new CSAssignment(Resources.createResource(0, 0), NodeType.NODE_LOCAL);
|
||||
|
||||
public static final CSAssignment SKIP_ASSIGNMENT = new CSAssignment(true);
|
||||
public static final CSAssignment SKIP_ASSIGNMENT =
|
||||
new CSAssignment(SkippedType.OTHER);
|
||||
|
||||
private Resource resource;
|
||||
private NodeType type;
|
||||
private RMContainer excessReservation;
|
||||
private FiCaSchedulerApp application;
|
||||
private final boolean skipped;
|
||||
private SkippedType skipped;
|
||||
|
||||
/**
|
||||
* Reason for the queue to get skipped.
|
||||
*/
|
||||
public enum SkippedType {
|
||||
NONE,
|
||||
QUEUE_LIMIT,
|
||||
OTHER
|
||||
}
|
||||
|
||||
private boolean fulfilledReservation;
|
||||
private final AssignmentInformation assignmentInformation;
|
||||
private boolean increaseAllocation;
|
||||
private List<RMContainer> containersToKill;
|
||||
|
||||
public CSAssignment(Resource resource, NodeType type) {
|
||||
this(resource, type, null, null, false, false);
|
||||
this(resource, type, null, null, SkippedType.NONE, false);
|
||||
}
|
||||
|
||||
public CSAssignment(FiCaSchedulerApp application,
|
||||
RMContainer excessReservation) {
|
||||
this(excessReservation.getContainer().getResource(), NodeType.NODE_LOCAL,
|
||||
excessReservation, application, false, false);
|
||||
excessReservation, application, SkippedType.NONE, false);
|
||||
}
|
||||
|
||||
public CSAssignment(boolean skipped) {
|
||||
public CSAssignment(SkippedType skipped) {
|
||||
this(Resource.newInstance(0, 0), NodeType.NODE_LOCAL, null, null, skipped,
|
||||
false);
|
||||
}
|
||||
|
||||
public CSAssignment(Resource resource, NodeType type,
|
||||
RMContainer excessReservation, FiCaSchedulerApp application,
|
||||
boolean skipped, boolean fulfilledReservation) {
|
||||
SkippedType skipped, boolean fulfilledReservation) {
|
||||
this.resource = resource;
|
||||
this.type = type;
|
||||
this.excessReservation = excessReservation;
|
||||
|
@ -105,10 +116,14 @@ public class CSAssignment {
|
|||
excessReservation = rmContainer;
|
||||
}
|
||||
|
||||
public boolean getSkipped() {
|
||||
public SkippedType getSkippedType() {
|
||||
return skipped;
|
||||
}
|
||||
|
||||
public void setSkippedType(SkippedType skippedType) {
|
||||
this.skipped = skippedType;
|
||||
}
|
||||
|
||||
@Override
|
||||
public String toString() {
|
||||
String ret = "resource:" + resource.toString();
|
||||
|
|
|
@ -60,6 +60,7 @@ import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainer;
|
|||
import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerEventType;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainerState;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.ActiveUsersManager;
|
||||
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.ResourceLimits;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.ResourceUsage;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedContainerChangeRequest;
|
||||
|
@ -972,8 +973,12 @@ public class LeafQueue extends AbstractCSQueue {
|
|||
|
||||
// Done
|
||||
return assignment;
|
||||
} else if (assignment.getSkipped()) {
|
||||
} else if (assignment.getSkippedType()
|
||||
== CSAssignment.SkippedType.OTHER) {
|
||||
application.updateNodeInfoForAMDiagnostics(node);
|
||||
} else if(assignment.getSkippedType()
|
||||
== CSAssignment.SkippedType.QUEUE_LIMIT) {
|
||||
return assignment;
|
||||
} else {
|
||||
// If we don't allocate anything, and it is not skipped by application,
|
||||
// we will return to respect FIFO of applications
|
||||
|
|
|
@ -37,7 +37,6 @@ import org.apache.hadoop.yarn.api.records.Resource;
|
|||
import org.apache.hadoop.yarn.exceptions.InvalidResourceRequestException;
|
||||
import org.apache.hadoop.yarn.factories.RecordFactory;
|
||||
import org.apache.hadoop.yarn.factory.providers.RecordFactoryProvider;
|
||||
import org.apache.hadoop.yarn.nodelabels.RMNodeLabel;
|
||||
import org.apache.hadoop.yarn.security.AccessType;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.nodelabels.RMNodeLabelsManager;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.rmcontainer.RMContainer;
|
||||
|
@ -474,6 +473,7 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
" cluster=" + clusterResource);
|
||||
|
||||
} else {
|
||||
assignment.setSkippedType(assignedToChild.getSkippedType());
|
||||
break;
|
||||
}
|
||||
|
||||
|
@ -511,14 +511,14 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
}
|
||||
|
||||
private ResourceLimits getResourceLimitsOfChild(CSQueue child,
|
||||
Resource clusterResource, ResourceLimits parentLimits,
|
||||
Resource clusterResource, Resource parentLimits,
|
||||
String nodePartition) {
|
||||
// Set resource-limit of a given child, child.limit =
|
||||
// min(my.limit - my.used + child.used, child.max)
|
||||
|
||||
// Parent available resource = parent-limit - parent-used-resource
|
||||
Resource parentMaxAvailableResource = Resources.subtract(
|
||||
parentLimits.getLimit(), queueUsage.getUsed(nodePartition));
|
||||
parentLimits, queueUsage.getUsed(nodePartition));
|
||||
// Deduct killable from used
|
||||
Resources.addTo(parentMaxAvailableResource,
|
||||
getTotalKillableResource(nodePartition));
|
||||
|
@ -568,9 +568,9 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
private synchronized CSAssignment assignContainersToChildQueues(
|
||||
Resource cluster, FiCaSchedulerNode node, ResourceLimits limits,
|
||||
SchedulingMode schedulingMode) {
|
||||
CSAssignment assignment =
|
||||
new CSAssignment(Resources.createResource(0, 0), NodeType.NODE_LOCAL);
|
||||
CSAssignment assignment = CSAssignment.NULL_ASSIGNMENT;
|
||||
|
||||
Resource parentLimits = limits.getLimit();
|
||||
printChildQueues();
|
||||
|
||||
// Try to assign to most 'under-served' sub-queue
|
||||
|
@ -584,20 +584,21 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
|
||||
// Get ResourceLimits of child queue before assign containers
|
||||
ResourceLimits childLimits =
|
||||
getResourceLimitsOfChild(childQueue, cluster, limits, node.getPartition());
|
||||
getResourceLimitsOfChild(childQueue, cluster, parentLimits,
|
||||
node.getPartition());
|
||||
|
||||
assignment = childQueue.assignContainers(cluster, node,
|
||||
CSAssignment childAssignment = childQueue.assignContainers(cluster, node,
|
||||
childLimits, schedulingMode);
|
||||
if(LOG.isDebugEnabled()) {
|
||||
LOG.debug("Assigned to queue: " + childQueue.getQueuePath() +
|
||||
" stats: " + childQueue + " --> " +
|
||||
assignment.getResource() + ", " + assignment.getType());
|
||||
" stats: " + childQueue + " --> " +
|
||||
childAssignment.getResource() + ", " + childAssignment.getType());
|
||||
}
|
||||
|
||||
// If we do assign, remove the queue and re-insert in-order to re-sort
|
||||
if (Resources.greaterThan(
|
||||
resourceCalculator, cluster,
|
||||
assignment.getResource(), Resources.none())) {
|
||||
childAssignment.getResource(), Resources.none())) {
|
||||
// Only update childQueues when we doing non-partitioned node
|
||||
// allocation.
|
||||
if (RMNodeLabelsManager.NO_LABEL.equals(node.getPartition())) {
|
||||
|
@ -610,7 +611,24 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
printChildQueues();
|
||||
}
|
||||
}
|
||||
assignment = childAssignment;
|
||||
break;
|
||||
} else if (childAssignment.getSkippedType() ==
|
||||
CSAssignment.SkippedType.QUEUE_LIMIT) {
|
||||
if (assignment.getSkippedType() !=
|
||||
CSAssignment.SkippedType.QUEUE_LIMIT) {
|
||||
assignment = childAssignment;
|
||||
}
|
||||
Resource resourceToSubtract = Resources.max(resourceCalculator,
|
||||
cluster, childLimits.getHeadroom(), Resources.none());
|
||||
if(LOG.isDebugEnabled()) {
|
||||
LOG.debug("Decrease parentLimits " + parentLimits +
|
||||
" for " + this.getQueueName() + " by " +
|
||||
resourceToSubtract + " as childQueue=" +
|
||||
childQueue.getQueueName() + " is blocked");
|
||||
}
|
||||
parentLimits = Resources.subtract(parentLimits,
|
||||
resourceToSubtract);
|
||||
}
|
||||
}
|
||||
|
||||
|
@ -731,7 +749,8 @@ public class ParentQueue extends AbstractCSQueue {
|
|||
for (CSQueue childQueue : childQueues) {
|
||||
// Get ResourceLimits of child queue before assign containers
|
||||
ResourceLimits childLimits = getResourceLimitsOfChild(childQueue,
|
||||
clusterResource, resourceLimits, RMNodeLabelsManager.NO_LABEL);
|
||||
clusterResource, resourceLimits.getLimit(),
|
||||
RMNodeLabelsManager.NO_LABEL);
|
||||
childQueue.updateClusterResource(clusterResource, childLimits);
|
||||
}
|
||||
|
||||
|
|
|
@ -55,8 +55,10 @@ public abstract class AbstractContainerAllocator {
|
|||
Resource clusterResource, ContainerAllocation result,
|
||||
RMContainer rmContainer) {
|
||||
// Handle skipped
|
||||
boolean skipped =
|
||||
(result.getAllocationState() == AllocationState.APP_SKIPPED);
|
||||
CSAssignment.SkippedType skipped =
|
||||
(result.getAllocationState() == AllocationState.APP_SKIPPED) ?
|
||||
CSAssignment.SkippedType.OTHER :
|
||||
CSAssignment.SkippedType.NONE;
|
||||
CSAssignment assignment = new CSAssignment(skipped);
|
||||
assignment.setApplication(application);
|
||||
|
||||
|
@ -110,6 +112,11 @@ public abstract class AbstractContainerAllocator {
|
|||
}
|
||||
|
||||
assignment.setContainersToKill(result.getToKillContainers());
|
||||
} else {
|
||||
if (result.getAllocationState() == AllocationState.QUEUE_SKIPPED) {
|
||||
assignment.setSkippedType(
|
||||
CSAssignment.SkippedType.QUEUE_LIMIT);
|
||||
}
|
||||
}
|
||||
|
||||
return assignment;
|
||||
|
|
|
@ -71,7 +71,7 @@ public class IncreaseContainerAllocator extends AbstractContainerAllocator {
|
|||
SchedContainerChangeRequest request) {
|
||||
CSAssignment assignment =
|
||||
new CSAssignment(request.getDeltaCapacity(), NodeType.NODE_LOCAL, null,
|
||||
application, false, false);
|
||||
application, CSAssignment.SkippedType.NONE, false);
|
||||
Resources.addTo(assignment.getAssignmentInformation().getReserved(),
|
||||
request.getDeltaCapacity());
|
||||
assignment.getAssignmentInformation().incrReservations();
|
||||
|
@ -88,7 +88,7 @@ public class IncreaseContainerAllocator extends AbstractContainerAllocator {
|
|||
SchedContainerChangeRequest request, boolean fromReservation) {
|
||||
CSAssignment assignment =
|
||||
new CSAssignment(request.getDeltaCapacity(), NodeType.NODE_LOCAL, null,
|
||||
application, false, fromReservation);
|
||||
application, CSAssignment.SkippedType.NONE, fromReservation);
|
||||
Resources.addTo(assignment.getAssignmentInformation().getAllocated(),
|
||||
request.getDeltaCapacity());
|
||||
assignment.getAssignmentInformation().incrAllocations();
|
||||
|
@ -311,7 +311,8 @@ public class IncreaseContainerAllocator extends AbstractContainerAllocator {
|
|||
// Try to allocate the increase request
|
||||
assigned =
|
||||
allocateIncreaseRequest(node, clusterResource, increaseRequest);
|
||||
if (!assigned.getSkipped()) {
|
||||
if (assigned.getSkippedType()
|
||||
== CSAssignment.SkippedType.NONE) {
|
||||
// When we don't skip this request, which means we either allocated
|
||||
// OR reserved this request. We will break
|
||||
break;
|
||||
|
@ -328,7 +329,8 @@ public class IncreaseContainerAllocator extends AbstractContainerAllocator {
|
|||
}
|
||||
|
||||
// We may have allocated something
|
||||
if (assigned != null && !assigned.getSkipped()) {
|
||||
if (assigned != null && assigned.getSkippedType()
|
||||
== CSAssignment.SkippedType.NONE) {
|
||||
break;
|
||||
}
|
||||
}
|
||||
|
|
|
@ -113,6 +113,8 @@ import org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNode;
|
|||
import org.apache.hadoop.yarn.server.resourcemanager.rmnode.RMNodeResourceUpdateEvent;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.AbstractYarnScheduler;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.Allocation;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.event.
|
||||
ContainerExpiredSchedulerEvent;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.ResourceScheduler;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedulerApplication;
|
||||
import org.apache.hadoop.yarn.server.resourcemanager.scheduler.SchedulerApplicationAttempt;
|
||||
|
@ -166,6 +168,12 @@ public class TestCapacityScheduler {
|
|||
private static final String B3 = B + ".b3";
|
||||
private static float A_CAPACITY = 10.5f;
|
||||
private static float B_CAPACITY = 89.5f;
|
||||
private static final String P1 = CapacitySchedulerConfiguration.ROOT + ".p1";
|
||||
private static final String P2 = CapacitySchedulerConfiguration.ROOT + ".p2";
|
||||
private static final String X1 = P1 + ".x1";
|
||||
private static final String X2 = P1 + ".x2";
|
||||
private static final String Y1 = P2 + ".y1";
|
||||
private static final String Y2 = P2 + ".y2";
|
||||
private static float A1_CAPACITY = 30;
|
||||
private static float A2_CAPACITY = 70;
|
||||
private static float B1_CAPACITY = 79.2f;
|
||||
|
@ -412,6 +420,51 @@ public class TestCapacityScheduler {
|
|||
return conf;
|
||||
}
|
||||
|
||||
private CapacitySchedulerConfiguration setupBlockedQueueConfiguration(
|
||||
CapacitySchedulerConfiguration conf) {
|
||||
|
||||
// Define top-level queues
|
||||
conf.setQueues(CapacitySchedulerConfiguration.ROOT,
|
||||
new String[]{"a", "b"});
|
||||
|
||||
conf.setCapacity(A, 80f);
|
||||
conf.setCapacity(B, 20f);
|
||||
conf.setUserLimitFactor(A, 100);
|
||||
conf.setUserLimitFactor(B, 100);
|
||||
conf.setMaximumCapacity(A, 100);
|
||||
conf.setMaximumCapacity(B, 100);
|
||||
LOG.info("Setup top-level queues a and b");
|
||||
return conf;
|
||||
}
|
||||
|
||||
private CapacitySchedulerConfiguration setupOtherBlockedQueueConfiguration(
|
||||
CapacitySchedulerConfiguration conf) {
|
||||
|
||||
// Define top-level queues
|
||||
conf.setQueues(CapacitySchedulerConfiguration.ROOT,
|
||||
new String[]{"p1", "p2"});
|
||||
|
||||
conf.setCapacity(P1, 50f);
|
||||
conf.setMaximumCapacity(P1, 50f);
|
||||
conf.setCapacity(P2, 50f);
|
||||
conf.setMaximumCapacity(P2, 100f);
|
||||
// Define 2nd-level queues
|
||||
conf.setQueues(P1, new String[] {"x1", "x2"});
|
||||
conf.setCapacity(X1, 80f);
|
||||
conf.setMaximumCapacity(X1, 100f);
|
||||
conf.setUserLimitFactor(X1, 2f);
|
||||
conf.setCapacity(X2, 20f);
|
||||
conf.setMaximumCapacity(X2, 100f);
|
||||
conf.setUserLimitFactor(X2, 2f);
|
||||
|
||||
conf.setQueues(P2, new String[]{"y1", "y2"});
|
||||
conf.setCapacity(Y1, 80f);
|
||||
conf.setUserLimitFactor(Y1, 2f);
|
||||
conf.setCapacity(Y2, 20f);
|
||||
conf.setUserLimitFactor(Y2, 2f);
|
||||
return conf;
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testMaximumCapacitySetup() {
|
||||
float delta = 0.0000001f;
|
||||
|
@ -3415,4 +3468,237 @@ public class TestCapacityScheduler {
|
|||
scheduler.handle(appRemovedEvent1);
|
||||
rm.stop();
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testCSReservationWithRootUnblocked() throws Exception {
|
||||
CapacitySchedulerConfiguration conf = new CapacitySchedulerConfiguration();
|
||||
conf.setResourceComparator(DominantResourceCalculator.class);
|
||||
setupOtherBlockedQueueConfiguration(conf);
|
||||
conf.setClass(YarnConfiguration.RM_SCHEDULER, CapacityScheduler.class,
|
||||
ResourceScheduler.class);
|
||||
MockRM rm = new MockRM(conf);
|
||||
rm.start();
|
||||
CapacityScheduler cs = (CapacityScheduler) rm.getResourceScheduler();
|
||||
ParentQueue q = (ParentQueue) cs.getQueue("p1");
|
||||
|
||||
Assert.assertNotNull(q);
|
||||
String host = "127.0.0.1";
|
||||
String host1 = "test";
|
||||
RMNode node =
|
||||
MockNodes.newNodeInfo(0, Resource.newInstance(8 * GB, 8), 1, host);
|
||||
RMNode node1 =
|
||||
MockNodes.newNodeInfo(0, Resource.newInstance(8 * GB, 8), 2, host1);
|
||||
cs.handle(new NodeAddedSchedulerEvent(node));
|
||||
cs.handle(new NodeAddedSchedulerEvent(node1));
|
||||
ApplicationAttemptId appAttemptId1 =
|
||||
appHelper(rm, cs, 100, 1, "x1", "userX1");
|
||||
ApplicationAttemptId appAttemptId2 =
|
||||
appHelper(rm, cs, 100, 2, "x2", "userX2");
|
||||
ApplicationAttemptId appAttemptId3 =
|
||||
appHelper(rm, cs, 100, 3, "y1", "userY1");
|
||||
RecordFactory recordFactory =
|
||||
RecordFactoryProvider.getRecordFactory(null);
|
||||
|
||||
Priority priority = TestUtils.createMockPriority(1);
|
||||
ResourceRequest y1Req = null;
|
||||
ResourceRequest x1Req = null;
|
||||
ResourceRequest x2Req = null;
|
||||
for(int i=0; i < 4; i++) {
|
||||
y1Req = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId3,
|
||||
Collections.<ResourceRequest>singletonList(y1Req),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
}
|
||||
assertEquals("Y1 Used Resource should be 4 GB", 4 * GB,
|
||||
cs.getQueue("y1").getUsedResources().getMemorySize());
|
||||
assertEquals("P2 Used Resource should be 4 GB", 4 * GB,
|
||||
cs.getQueue("p2").getUsedResources().getMemorySize());
|
||||
|
||||
for(int i=0; i < 7; i++) {
|
||||
x1Req = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId1,
|
||||
Collections.<ResourceRequest>singletonList(x1Req),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
}
|
||||
assertEquals("X1 Used Resource should be 7 GB", 7 * GB,
|
||||
cs.getQueue("x1").getUsedResources().getMemorySize());
|
||||
assertEquals("P1 Used Resource should be 7 GB", 7 * GB,
|
||||
cs.getQueue("p1").getUsedResources().getMemorySize());
|
||||
|
||||
x2Req = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 2 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId2,
|
||||
Collections.<ResourceRequest>singletonList(x2Req),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
assertEquals("X2 Used Resource should be 0", 0,
|
||||
cs.getQueue("x2").getUsedResources().getMemorySize());
|
||||
assertEquals("P1 Used Resource should be 7 GB", 7 * GB,
|
||||
cs.getQueue("p1").getUsedResources().getMemorySize());
|
||||
//this assign should fail
|
||||
x1Req = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId1,
|
||||
Collections.<ResourceRequest>singletonList(x1Req),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
assertEquals("X1 Used Resource should be 7 GB", 7 * GB,
|
||||
cs.getQueue("x1").getUsedResources().getMemorySize());
|
||||
assertEquals("P1 Used Resource should be 7 GB", 7 * GB,
|
||||
cs.getQueue("p1").getUsedResources().getMemorySize());
|
||||
|
||||
//this should get thru
|
||||
for (int i=0; i < 4; i++) {
|
||||
y1Req = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId3,
|
||||
Collections.<ResourceRequest>singletonList(y1Req),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
}
|
||||
assertEquals("P2 Used Resource should be 8 GB", 8 * GB,
|
||||
cs.getQueue("p2").getUsedResources().getMemorySize());
|
||||
|
||||
//Free a container from X1
|
||||
ContainerId containerId = ContainerId.newContainerId(appAttemptId1, 2);
|
||||
cs.handle(new ContainerExpiredSchedulerEvent(containerId));
|
||||
|
||||
//Schedule pending request
|
||||
CapacityScheduler.schedule(cs);
|
||||
assertEquals("X2 Used Resource should be 2 GB", 2 * GB,
|
||||
cs.getQueue("x2").getUsedResources().getMemorySize());
|
||||
assertEquals("P1 Used Resource should be 8 GB", 8 * GB,
|
||||
cs.getQueue("p1").getUsedResources().getMemorySize());
|
||||
assertEquals("P2 Used Resource should be 8 GB", 8 * GB,
|
||||
cs.getQueue("p2").getUsedResources().getMemorySize());
|
||||
assertEquals("Root Used Resource should be 16 GB", 16 * GB,
|
||||
cs.getRootQueue().getUsedResources().getMemorySize());
|
||||
rm.stop();
|
||||
}
|
||||
|
||||
@Test
|
||||
public void testCSQueueBlocked() throws Exception {
|
||||
CapacitySchedulerConfiguration conf = new CapacitySchedulerConfiguration();
|
||||
setupBlockedQueueConfiguration(conf);
|
||||
conf.setClass(YarnConfiguration.RM_SCHEDULER, CapacityScheduler.class,
|
||||
ResourceScheduler.class);
|
||||
MockRM rm = new MockRM(conf);
|
||||
rm.start();
|
||||
CapacityScheduler cs = (CapacityScheduler) rm.getResourceScheduler();
|
||||
LeafQueue q = (LeafQueue) cs.getQueue("a");
|
||||
|
||||
Assert.assertNotNull(q);
|
||||
String host = "127.0.0.1";
|
||||
String host1 = "test";
|
||||
RMNode node =
|
||||
MockNodes.newNodeInfo(0, Resource.newInstance(8 * GB, 8), 1, host);
|
||||
RMNode node1 =
|
||||
MockNodes.newNodeInfo(0, Resource.newInstance(8 * GB, 8), 2, host1);
|
||||
cs.handle(new NodeAddedSchedulerEvent(node));
|
||||
cs.handle(new NodeAddedSchedulerEvent(node1));
|
||||
//add app begin
|
||||
ApplicationAttemptId appAttemptId1 =
|
||||
appHelper(rm, cs, 100, 1, "a", "user1");
|
||||
ApplicationAttemptId appAttemptId2 =
|
||||
appHelper(rm, cs, 100, 2, "b", "user2");
|
||||
//add app end
|
||||
|
||||
RecordFactory recordFactory =
|
||||
RecordFactoryProvider.getRecordFactory(null);
|
||||
|
||||
Priority priority = TestUtils.createMockPriority(1);
|
||||
ResourceRequest r1 = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 2 * GB, 1, true, priority, recordFactory);
|
||||
//This will allocate for app1
|
||||
cs.allocate(appAttemptId1, Collections.<ResourceRequest>singletonList(r1),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null).getContainers().size();
|
||||
CapacityScheduler.schedule(cs);
|
||||
ResourceRequest r2 = null;
|
||||
for (int i =0; i < 13; i++) {
|
||||
r2 = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId2,
|
||||
Collections.<ResourceRequest>singletonList(r2),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
}
|
||||
assertEquals("A Used Resource should be 2 GB", 2 * GB,
|
||||
cs.getQueue("a").getUsedResources().getMemorySize());
|
||||
assertEquals("B Used Resource should be 2 GB", 13 * GB,
|
||||
cs.getQueue("b").getUsedResources().getMemorySize());
|
||||
r1 = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 2 * GB, 1, true, priority, recordFactory);
|
||||
r2 = TestUtils.createResourceRequest(
|
||||
ResourceRequest.ANY, 1 * GB, 1, true, priority, recordFactory);
|
||||
cs.allocate(appAttemptId1, Collections.<ResourceRequest>singletonList(r1),
|
||||
Collections.<ContainerId>emptyList(),
|
||||
null, null, null, null).getContainers().size();
|
||||
CapacityScheduler.schedule(cs);
|
||||
|
||||
cs.allocate(appAttemptId2, Collections.<ResourceRequest>singletonList(r2),
|
||||
Collections.<ContainerId>emptyList(), null, null, null, null);
|
||||
CapacityScheduler.schedule(cs);
|
||||
//Check blocked Resource
|
||||
assertEquals("A Used Resource should be 2 GB", 2 * GB,
|
||||
cs.getQueue("a").getUsedResources().getMemorySize());
|
||||
assertEquals("B Used Resource should be 13 GB", 13 * GB,
|
||||
cs.getQueue("b").getUsedResources().getMemorySize());
|
||||
|
||||
ContainerId containerId1 = ContainerId.newContainerId(appAttemptId2, 10);
|
||||
ContainerId containerId2 =ContainerId.newContainerId(appAttemptId2, 11);
|
||||
|
||||
cs.handle(new ContainerExpiredSchedulerEvent(containerId1));
|
||||
cs.handle(new ContainerExpiredSchedulerEvent(containerId2));
|
||||
CapacityScheduler.schedule(cs);
|
||||
rm.drainEvents();
|
||||
assertEquals("A Used Resource should be 2 GB", 4 * GB,
|
||||
cs.getQueue("a").getUsedResources().getMemorySize());
|
||||
assertEquals("B Used Resource should be 12 GB", 12 * GB,
|
||||
cs.getQueue("b").getUsedResources().getMemorySize());
|
||||
assertEquals("Used Resource on Root should be 16 GB", 16 * GB,
|
||||
cs.getRootQueue().getUsedResources().getMemorySize());
|
||||
rm.stop();
|
||||
}
|
||||
|
||||
private ApplicationAttemptId appHelper(MockRM rm, CapacityScheduler cs,
|
||||
int clusterTs, int appId, String queue,
|
||||
String user) {
|
||||
ApplicationId appId1 = BuilderUtils.newApplicationId(clusterTs, appId);
|
||||
ApplicationAttemptId appAttemptId1 = BuilderUtils.newApplicationAttemptId(
|
||||
appId1, appId);
|
||||
|
||||
RMAppAttemptMetrics attemptMetric1 =
|
||||
new RMAppAttemptMetrics(appAttemptId1, rm.getRMContext());
|
||||
RMAppImpl app1 = mock(RMAppImpl.class);
|
||||
when(app1.getApplicationId()).thenReturn(appId1);
|
||||
RMAppAttemptImpl attempt1 = mock(RMAppAttemptImpl.class);
|
||||
Container container = mock(Container.class);
|
||||
when(attempt1.getMasterContainer()).thenReturn(container);
|
||||
ApplicationSubmissionContext submissionContext = mock(
|
||||
ApplicationSubmissionContext.class);
|
||||
when(attempt1.getSubmissionContext()).thenReturn(submissionContext);
|
||||
when(attempt1.getAppAttemptId()).thenReturn(appAttemptId1);
|
||||
when(attempt1.getRMAppAttemptMetrics()).thenReturn(attemptMetric1);
|
||||
when(app1.getCurrentAppAttempt()).thenReturn(attempt1);
|
||||
rm.getRMContext().getRMApps().put(appId1, app1);
|
||||
|
||||
SchedulerEvent addAppEvent1 =
|
||||
new AppAddedSchedulerEvent(appId1, queue, user);
|
||||
cs.handle(addAppEvent1);
|
||||
SchedulerEvent addAttemptEvent1 =
|
||||
new AppAttemptAddedSchedulerEvent(appAttemptId1, false);
|
||||
cs.handle(addAttemptEvent1);
|
||||
return appAttemptId1;
|
||||
}
|
||||
}
|
||||
|
|
Loading…
Reference in New Issue