Apache
/
hbase
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

HBASE-27125 The batch size of cleaning expired mob files should have an upper bound (#4541) 

Signed-off-by: Duo Zhang <zhangduo@apache.org>
This commit is contained in:
Xiaolin Ha 2022-06-17 23:35:26 +08:00 committed by GitHub
parent 5e5d18bf21
commit 89bda2c322
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
3 changed files with 47 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

3
hbase-server/src/main/java/org/apache/hadoop/hbase/mob/MobConstants.java
View File

@ -46,6 +46,9 @@ public final class MobConstants {
public static final String MOB_CACHE_EVICT_REMAIN_RATIO = "hbase.mob.cache.evict.remain.ratio";
public static final Tag MOB_REF_TAG =
new ArrayBackedTag(TagType.MOB_REFERENCE_TAG_TYPE, HConstants.EMPTY_BYTE_ARRAY);
public static final String MOB_CLEANER_BATCH_SIZE_UPPER_BOUND =
"hbase.master.mob.cleaner.batch.size.upper.bound";
public static final int DEFAULT_MOB_CLEANER_BATCH_SIZE_UPPER_BOUND = 10000;
public static final float DEFAULT_EVICT_REMAIN_RATIO = 0.5f;
public static final long DEFAULT_MOB_CACHE_EVICT_PERIOD = 3600L;

44
hbase-server/src/main/java/org/apache/hadoop/hbase/mob/MobUtils.java
View File

@ -17,6 +17,9 @@
*/
package org.apache.hadoop.hbase.mob;
import static org.apache.hadoop.hbase.mob.MobConstants.DEFAULT_MOB_CLEANER_BATCH_SIZE_UPPER_BOUND;
import static org.apache.hadoop.hbase.mob.MobConstants.MOB_CLEANER_BATCH_SIZE_UPPER_BOUND;
import java.io.FileNotFoundException;
import java.io.IOException;
import java.nio.ByteBuffer;
@ -314,20 +317,30 @@ public final class MobUtils {
}
filesToClean
.add(new HStoreFile(fs, file.getPath(), conf, cacheConfig, BloomType.NONE, true));
if (
filesToClean.size() >= conf.getInt(MOB_CLEANER_BATCH_SIZE_UPPER_BOUND,
DEFAULT_MOB_CLEANER_BATCH_SIZE_UPPER_BOUND)
) {
if (
removeMobFiles(conf, fs, tableName, mobTableDir, columnDescriptor.getName(),
filesToClean)
) {
deletedFileCount += filesToClean.size();
}
filesToClean.clear();
}
}
} catch (Exception e) {
LOG.error("Cannot parse the fileName " + fileName, e);
}
}
if (!filesToClean.isEmpty()) {
try {
removeMobFiles(conf, fs, tableName, mobTableDir, columnDescriptor.getName(), filesToClean);
deletedFileCount = filesToClean.size();
} catch (IOException e) {
LOG.error("Failed to delete the mob files " + filesToClean, e);
}
if (
!filesToClean.isEmpty() && removeMobFiles(conf, fs, tableName, mobTableDir,
columnDescriptor.getName(), filesToClean)
) {
deletedFileCount += filesToClean.size();
}
LOG.info("{} expired mob files are deleted", deletedFileCount);
LOG.info("Table {} {} expired mob files in total are deleted", tableName, deletedFileCount);
}
/**
@ -459,10 +472,17 @@ public final class MobUtils {
* @param family The name of the column family.
* @param storeFiles The files to be deleted.
*/
public static void removeMobFiles(Configuration conf, FileSystem fs, TableName tableName,
Path tableDir, byte[] family, Collection<HStoreFile> storeFiles) throws IOException {
HFileArchiver.archiveStoreFiles(conf, fs, getMobRegionInfo(tableName), tableDir, family,
storeFiles);
public static boolean removeMobFiles(Configuration conf, FileSystem fs, TableName tableName,
Path tableDir, byte[] family, Collection<HStoreFile> storeFiles) {
try {
HFileArchiver.archiveStoreFiles(conf, fs, getMobRegionInfo(tableName), tableDir, family,
storeFiles);
LOG.info("Table {} {} expired mob files are deleted", tableName, storeFiles.size());
return true;
} catch (IOException e) {
LOG.error("Failed to delete the mob files, table {}", tableName, e);
}
return false;
}
/**

13
hbase-server/src/test/java/org/apache/hadoop/hbase/mob/TestExpiredMobFileCleaner.java
View File

@ -17,6 +17,7 @@
*/
package org.apache.hadoop.hbase.mob;
import static org.apache.hadoop.hbase.mob.MobConstants.MOB_CLEANER_BATCH_SIZE_UPPER_BOUND;
import static org.junit.Assert.assertEquals;
import org.apache.hadoop.fs.FileStatus;
@ -55,6 +56,7 @@ public class TestExpiredMobFileCleaner {
private final static String family = "family";
private final static byte[] row1 = Bytes.toBytes("row1");
private final static byte[] row2 = Bytes.toBytes("row2");
private final static byte[] row3 = Bytes.toBytes("row3");
private final static byte[] qf = Bytes.toBytes("qf");
private static BufferedMutator table;
@ -63,6 +65,7 @@ public class TestExpiredMobFileCleaner {
@BeforeClass
public static void setUpBeforeClass() throws Exception {
TEST_UTIL.getConfiguration().setInt("hfile.format.version", 3);
TEST_UTIL.getConfiguration().setInt(MOB_CLEANER_BATCH_SIZE_UPPER_BOUND, 2);
}
@AfterClass
@ -146,6 +149,14 @@ public class TestExpiredMobFileCleaner {
String f2 = secondFiles[1].getPath().getName();
String secondFile = f1.equals(firstFile) ? f2 : f1;
ts = EnvironmentEdgeManager.currentTime() - 4 * secondsOfDay() * 1000; // 4 days before
putKVAndFlush(table, row3, dummyData, ts);
ts = EnvironmentEdgeManager.currentTime() - 4 * secondsOfDay() * 1000; // 4 days before
putKVAndFlush(table, row3, dummyData, ts);
FileStatus[] thirdFiles = TEST_UTIL.getTestFileSystem().listStatus(mobDirPath);
// now there are 4 mob files
assertEquals("Before cleanup without delay 3", 4, thirdFiles.length);
modifyColumnExpiryDays(2); // ttl = 2, make the first row expired
// run the cleaner
@ -156,7 +167,7 @@ public class TestExpiredMobFileCleaner {
FileStatus[] filesAfterClean = TEST_UTIL.getTestFileSystem().listStatus(mobDirPath);
String lastFile = filesAfterClean[0].getPath().getName();
// the first mob fie is removed
// there are 4 mob files in total, but only 3 need to be cleaned
assertEquals("After cleanup without delay 1", 1, filesAfterClean.length);
assertEquals("After cleanup without delay 2", secondFile, lastFile);
}