Class TestLogRollAbort

java.lang.Object
org.apache.hadoop.hbase.regionserver.wal.TestLogRollAbort

@Tag("org.apache.hadoop.hbase.testclassification.RegionServerTests") @Tag("org.apache.hadoop.hbase.testclassification.MediumTests") public class TestLogRollAbort extends Object
Tests for conditions that should trigger RegionServer aborts when rolling the current WAL fails.
  • Field Details

    • LOG

      private static final org.slf4j.Logger LOG
    • dfsCluster

      private static org.apache.hadoop.hdfs.MiniDFSCluster dfsCluster
    • admin

      private static org.apache.hadoop.hbase.client.Admin admin
    • cluster

    • TEST_UTIL

      protected static final HBaseTestingUtil TEST_UTIL
    • HBASEDIR

      private static final org.apache.hadoop.fs.Path HBASEDIR
    • HBASELOGDIR

      private static final org.apache.hadoop.fs.Path HBASELOGDIR
    • OLDLOGDIR

      private static final org.apache.hadoop.fs.Path OLDLOGDIR
    • conf

      private org.apache.hadoop.conf.Configuration conf
    • fs

      private org.apache.hadoop.fs.FileSystem fs
    • walProvider

      private final String walProvider
  • Constructor Details

  • Method Details

    • setUpBeforeClass

      @BeforeAll public static void setUpBeforeClass() throws Exception
      Throws:
      Exception
    • parameters

      public static Stream<org.junit.jupiter.params.provider.Arguments> parameters()
    • setUp

      @BeforeEach public void setUp() throws Exception
      Throws:
      Exception
    • tearDown

      @AfterEach public void tearDown() throws Exception
      Throws:
      Exception
    • testRSAbortWithUnflushedEdits

      @TestTemplate public void testRSAbortWithUnflushedEdits() throws Exception
      Tests that RegionServer aborts if we hit an error closing the WAL when there are unsynced WAL edits. See HBASE-4282.
      Throws:
      Exception
    • testLogRollAfterSplitStart

      @TestTemplate public void testLogRollAfterSplitStart() throws IOException
      Tests the case where a RegionServer enters a GC pause, comes back online after the master declared it dead and started to split. Want log rolling after a master split to fail. See HBASE-2312.
      Throws:
      IOException
    • closeWriter

      private <W extends org.apache.hadoop.hbase.wal.WALProvider.WriterBase> void closeWriter(org.apache.hadoop.hbase.regionserver.wal.AbstractFSWAL<W> wal)