org.apache.hadoop.hbase.snapshot

Class SnapshotManifest

java.lang.Object

org.apache.hadoop.hbase.snapshot.SnapshotManifest

@InterfaceAudience.Private
public class SnapshotManifest
extends Object

Utility class to help read/write the Snapshot Manifest. The snapshot format is transparent for the users of this class, once the snapshot is written, it will never be modified. On open() the snapshot will be loaded to the current in-memory format.

Field Summary

Fields

Modifier and Type	Field and Description
static String	DATA_MANIFEST_NAME
static String	SNAPSHOT_MANIFEST_SIZE_LIMIT_CONF_KEY

Method Summary

Methods

Modifier and Type	Method and Description
void	addRegion(HRegion region) Creates a 'manifest' for the specified region, by reading directly from the HRegion object.
void	addRegion(org.apache.hadoop.fs.Path tableDir, HRegionInfo regionInfo) Creates a 'manifest' for the specified region, by reading directly from the disk.
void	addTableDescriptor(HTableDescriptor htd) Add the table descriptor to the snapshot manifest
void	consolidate()
static SnapshotManifest	create(org.apache.hadoop.conf.Configuration conf, org.apache.hadoop.fs.FileSystem fs, org.apache.hadoop.fs.Path workingDir, HBaseProtos.SnapshotDescription desc, ForeignExceptionSnare monitor) Return a SnapshotManifest instance, used for writing a snapshot.
static ThreadPoolExecutor	createExecutor(org.apache.hadoop.conf.Configuration conf, String name)
List<SnapshotProtos.SnapshotRegionManifest>	getRegionManifests() Get all the Region Manifest from the snapshot
Map<String,SnapshotProtos.SnapshotRegionManifest>	getRegionManifestsMap() Get all the Region Manifest from the snapshot.
HBaseProtos.SnapshotDescription	getSnapshotDescription() Get the SnapshotDescription
org.apache.hadoop.fs.Path	getSnapshotDir() Get the current snapshot working dir
HTableDescriptor	getTableDescriptor() Get the table descriptor from the Snapshot
static SnapshotManifest	open(org.apache.hadoop.conf.Configuration conf, org.apache.hadoop.fs.FileSystem fs, org.apache.hadoop.fs.Path workingDir, HBaseProtos.SnapshotDescription desc) Return a SnapshotManifest instance with the information already loaded in-memory.

Methods inherited from class java.lang.Object

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

Field Detail

SNAPSHOT_MANIFEST_SIZE_LIMIT_CONF_KEY

public static final String SNAPSHOT_MANIFEST_SIZE_LIMIT_CONF_KEY

See Also:

Constant Field Values

DATA_MANIFEST_NAME

public static final String DATA_MANIFEST_NAME

See Also:

Constant Field Values

Method Detail

create

public static SnapshotManifest create(org.apache.hadoop.conf.Configuration conf,
                      org.apache.hadoop.fs.FileSystem fs,
                      org.apache.hadoop.fs.Path workingDir,
                      HBaseProtos.SnapshotDescription desc,
                      ForeignExceptionSnare monitor)

Return a SnapshotManifest instance, used for writing a snapshot. There are two usage pattern: - The Master will create a manifest, add the descriptor, offline regions and consolidate the snapshot by writing all the pending stuff on-disk. manifest = SnapshotManifest.create(...) manifest.addRegion(tableDir, hri) manifest.consolidate() - The RegionServer will create a single region manifest manifest = SnapshotManifest.create(...) manifest.addRegion(region)

open

public static SnapshotManifest open(org.apache.hadoop.conf.Configuration conf,
                    org.apache.hadoop.fs.FileSystem fs,
                    org.apache.hadoop.fs.Path workingDir,
                    HBaseProtos.SnapshotDescription desc)
                             throws IOException

Return a SnapshotManifest instance with the information already loaded in-memory. SnapshotManifest manifest = SnapshotManifest.open(...) HTableDescriptor htd = manifest.getTableDescriptor() for (SnapshotRegionManifest regionManifest: manifest.getRegionManifests()) hri = regionManifest.getRegionInfo() for (regionManifest.getFamilyFiles()) ...

Throws:

IOException

addTableDescriptor

public void addTableDescriptor(HTableDescriptor htd)
                        throws IOException

Add the table descriptor to the snapshot manifest

Throws:

IOException

addRegion

public void addRegion(HRegion region)
               throws IOException

Creates a 'manifest' for the specified region, by reading directly from the HRegion object. This is used by the "online snapshot" when the table is enabled.

Throws:

IOException

addRegion

public void addRegion(org.apache.hadoop.fs.Path tableDir,
             HRegionInfo regionInfo)
               throws IOException

Creates a 'manifest' for the specified region, by reading directly from the disk. This is used by the "offline snapshot" when the table is disabled.

Throws:

IOException

getSnapshotDir

public org.apache.hadoop.fs.Path getSnapshotDir()

Get the current snapshot working dir

getSnapshotDescription

public HBaseProtos.SnapshotDescription getSnapshotDescription()

Get the SnapshotDescription

getTableDescriptor

public HTableDescriptor getTableDescriptor()

Get the table descriptor from the Snapshot

getRegionManifests

public List<SnapshotProtos.SnapshotRegionManifest> getRegionManifests()

Get all the Region Manifest from the snapshot

getRegionManifestsMap

public Map<String,SnapshotProtos.SnapshotRegionManifest> getRegionManifestsMap()

Get all the Region Manifest from the snapshot. This is an helper to get a map with the region encoded name

consolidate

public void consolidate()
                 throws IOException

Throws:

IOException

createExecutor

public static ThreadPoolExecutor createExecutor(org.apache.hadoop.conf.Configuration conf,
                                String name)