org.apache.hadoop.hbase.backup.example

Class ZKTableArchiveClient

java.lang.Object

org.apache.hadoop.conf.Configured

org.apache.hadoop.hbase.backup.example.ZKTableArchiveClient

All Implemented Interfaces:

org.apache.hadoop.conf.Configurable

@InterfaceAudience.Private
public class ZKTableArchiveClient
extends org.apache.hadoop.conf.Configured

Example class for how to use the table archiving coordinated via zookeeper

Constructor Summary

Constructors

Constructor and Description
ZKTableArchiveClient(org.apache.hadoop.conf.Configuration conf, ClusterConnection connection)

Method Summary

Methods

Modifier and Type	Method and Description
void	disableHFileBackup() Disable hfile backups for all tables.
void	disableHFileBackup(byte[] table) Disable hfile backups for the given table.
void	disableHFileBackup(String table) Disable hfile backups for the given table.
void	enableHFileBackupAsync(byte[] table) Turn on backups for all HFiles for the given table.
static String	getArchiveZNode(org.apache.hadoop.conf.Configuration conf, ZooKeeperWatcher zooKeeper)
boolean	getArchivingEnabled(byte[] table) Determine if archiving is enabled (but not necessarily fully propagated) for a table
boolean	getArchivingEnabled(String table) Determine if archiving is enabled (but not necessarily fully propagated) for a table

Methods inherited from class org.apache.hadoop.conf.Configured

getConf, setConf

Methods inherited from class java.lang.Object

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

Constructor Detail

ZKTableArchiveClient

public ZKTableArchiveClient(org.apache.hadoop.conf.Configuration conf,
                    ClusterConnection connection)

Method Detail

enableHFileBackupAsync

public void enableHFileBackupAsync(byte[] table)
                            throws IOException,
                                   org.apache.zookeeper.KeeperException

Turn on backups for all HFiles for the given table.

All deleted hfiles are moved to the archive directory under the table directory, rather than being deleted.

If backups are already enabled for this table, does nothing.

If the table does not exist, the archiving the table's hfiles is still enabled as a future table with that name may be created shortly.

Parameters:

table - name of the table to start backing up

Throws:

IOException - if an unexpected exception occurs

org.apache.zookeeper.KeeperException - if zookeeper can't be reached

disableHFileBackup

public void disableHFileBackup(String table)
                        throws IOException,
                               org.apache.zookeeper.KeeperException

Disable hfile backups for the given table.

Previously backed up files are still retained (if present).

Asynchronous operation - some extra HFiles may be retained, in the archive directory after disable is called, dependent on the latency in zookeeper to the servers.

Parameters:

table - name of the table stop backing up

Throws:

IOException - if an unexpected exception occurs

org.apache.zookeeper.KeeperException - if zookeeper can't be reached

disableHFileBackup

public void disableHFileBackup(byte[] table)
                        throws IOException,
                               org.apache.zookeeper.KeeperException

Disable hfile backups for the given table.

Previously backed up files are still retained (if present).

Asynchronous operation - some extra HFiles may be retained, in the archive directory after disable is called, dependent on the latency in zookeeper to the servers.

Parameters:

table - name of the table stop backing up

Throws:

IOException - if an unexpected exception occurs

org.apache.zookeeper.KeeperException - if zookeeper can't be reached

disableHFileBackup

public void disableHFileBackup()
                        throws IOException,
                               org.apache.zookeeper.KeeperException

Disable hfile backups for all tables.

Previously backed up files are still retained (if present).

Asynchronous operation - some extra HFiles may be retained, in the archive directory after disable is called, dependent on the latency in zookeeper to the servers.

Throws:

IOException - if an unexpected exception occurs

org.apache.zookeeper.KeeperException - if zookeeper can't be reached

getArchivingEnabled

public boolean getArchivingEnabled(byte[] table)
                            throws IOException,
                                   org.apache.zookeeper.KeeperException

Determine if archiving is enabled (but not necessarily fully propagated) for a table

Parameters:

table - name of the table to check

Returns:

true if it is, false otherwise

Throws:

IOException - if a connection to ZooKeeper cannot be established

org.apache.zookeeper.KeeperException

getArchivingEnabled

public boolean getArchivingEnabled(String table)
                            throws IOException,
                                   org.apache.zookeeper.KeeperException

Determine if archiving is enabled (but not necessarily fully propagated) for a table

Parameters:

table - name of the table to check

Returns:

true if it is, false otherwise

Throws:

IOException - if an unexpected network issue occurs

org.apache.zookeeper.KeeperException - if zookeeper can't be reached

getArchiveZNode

public static String getArchiveZNode(org.apache.hadoop.conf.Configuration conf,
                     ZooKeeperWatcher zooKeeper)

Parameters:

conf - conf to read for the base archive node

zooKeeper - zookeeper to used for building the full path

Returns:

get the znode for long-term archival of a table for