Package org.apache.hadoop.hbase.client

Class TestRegionLocatorPagedScanRpcCount

java.lang.Object

org.apache.hadoop.hbase.client.TestRegionLocatorPagedScanRpcCount

@Tag("org.apache.hadoop.hbase.testclassification.MediumTests") @Tag("org.apache.hadoop.hbase.testclassification.ClientTests")
public class TestRegionLocatorPagedScanRpcCount
extends Object

Asserts the single-RPC promise of the paginated meta-scan path (ClientMetaTableAccessor.getTableHRegionLocations(AsyncTable, TableName, byte[], int)) by wrapping the meta AsyncTable so we can count AdvancedScanResultConsumer.onNext(org.apache.hadoop.hbase.client.Result[], org.apache.hadoop.hbase.client.AdvancedScanResultConsumer.ScanController) invocations - one per ScannerNext server response.

Cluster runs with hbase.meta.scanner.caching = 2 so the limit > caching branch is exercised cheaply with a small table (5 user regions).

Field Summary

Fields

Modifier and Type	Field	Description
private static org.apache.hadoop.hbase.client.AsyncConnection	CONN
private static final byte[]	FAMILY
private static final int	META_CACHING	Caching is small enough that limit > META_CACHING is easy to set up.
private static final int	NUM_REGIONS
private static final org.apache.hadoop.hbase.TableName	TABLE_NAME
private static final HBaseTestingUtil	UTIL

Constructor Summary

Constructors

Constructor	Description
TestRegionLocatorPagedScanRpcCount()

Method Summary

Modifier and Type	Method	Description
private int	runPagedScanAndCountRpcs(int limit)
static void	setUp()
static void	tearDown()
void	testSingleRpcWhenLimitExceedsCaching()
void	testSingleRpcWhenLimitWithinCaching()
void	testUnboundedPathStillUsesConfiguredCaching()
private static org.apache.hadoop.hbase.client.AsyncTable<org.apache.hadoop.hbase.client.AdvancedScanResultConsumer>	wrapMetaTable(AtomicInteger onNextCalls)	Returns a delegating proxy for the meta AsyncTable that intercepts scan(Scan, AdvancedScanResultConsumer) and wraps the supplied consumer so every AdvancedScanResultConsumer.onNext(org.apache.hadoop.hbase.client.Result[], org.apache.hadoop.hbase.client.AdvancedScanResultConsumer.ScanController) invocation increments onNextCalls.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Details

UTIL

private static final HBaseTestingUtil UTIL

TABLE_NAME

private static final org.apache.hadoop.hbase.TableName TABLE_NAME

FAMILY

private static final byte[] FAMILY

META_CACHING

private static final int META_CACHING

Caching is small enough that limit > META_CACHING is easy to set up.

See Also:

• Constant Field Values

NUM_REGIONS

private static final int NUM_REGIONS

See Also:

• Constant Field Values

CONN

private static org.apache.hadoop.hbase.client.AsyncConnection CONN

Constructor Details

TestRegionLocatorPagedScanRpcCount

public TestRegionLocatorPagedScanRpcCount()

Method Details

setUp

@BeforeAll
public static void setUp()
                  throws Exception

Throws:

Exception

tearDown

@AfterAll
public static void tearDown()
                     throws Exception

Throws:

Exception

testSingleRpcWhenLimitWithinCaching

@Test
public void testSingleRpcWhenLimitWithinCaching()
                                         throws Exception

Throws:

Exception

testSingleRpcWhenLimitExceedsCaching

@Test
public void testSingleRpcWhenLimitExceedsCaching()
                                          throws Exception

Throws:

Exception

testUnboundedPathStillUsesConfiguredCaching

@Test
public void testUnboundedPathStillUsesConfiguredCaching()
                                                 throws Exception

Throws:

Exception

runPagedScanAndCountRpcs

private int runPagedScanAndCountRpcs(int limit)
                              throws Exception

Throws:

Exception

wrapMetaTable

private static org.apache.hadoop.hbase.client.AsyncTable<org.apache.hadoop.hbase.client.AdvancedScanResultConsumer> wrapMetaTable(AtomicInteger onNextCalls)

Returns a delegating proxy for the meta AsyncTable that intercepts scan(Scan, AdvancedScanResultConsumer) and wraps the supplied consumer so every AdvancedScanResultConsumer.onNext(org.apache.hadoop.hbase.client.Result[], org.apache.hadoop.hbase.client.AdvancedScanResultConsumer.ScanController) invocation increments onNextCalls.