本文整理汇总了Java中org.apache.hadoop.hbase.HTableDescriptor.getOwnerString方法的典型用法代码示例。如果您正苦于以下问题:Java HTableDescriptor.getOwnerString方法的具体用法?Java HTableDescriptor.getOwnerString怎么用?Java HTableDescriptor.getOwnerString使用的例子?那么恭喜您, 这里精选的方法代码示例或许可以为您提供帮助。您也可以进一步了解该方法所在类org.apache.hadoop.hbase.HTableDescriptor
的用法示例。
在下文中一共展示了HTableDescriptor.getOwnerString方法的2个代码示例,这些例子默认根据受欢迎程度排序。您可以为喜欢或者感觉有用的代码点赞,您的评价将有助于系统推荐出更棒的Java代码示例。
示例1: postModifyTable
import org.apache.hadoop.hbase.HTableDescriptor; //导入方法依赖的package包/类
@Override
public void postModifyTable(ObserverContext<MasterCoprocessorEnvironment> c,
TableName tableName, final HTableDescriptor htd) throws IOException {
final Configuration conf = c.getEnvironment().getConfiguration();
// default the table owner to current user, if not specified.
final String owner = (htd.getOwnerString() != null) ? htd.getOwnerString() :
getActiveUser().getShortName();
User.runAsLoginUser(new PrivilegedExceptionAction<Void>() {
@Override
public Void run() throws Exception {
UserPermission userperm = new UserPermission(Bytes.toBytes(owner),
htd.getTableName(), null, Action.values());
AccessControlLists.addUserPermission(conf, userperm);
return null;
}
});
}
示例2: postCreateTableHandler
import org.apache.hadoop.hbase.HTableDescriptor; //导入方法依赖的package包/类
@Override
public void postCreateTableHandler(final ObserverContext<MasterCoprocessorEnvironment> c,
HTableDescriptor desc, HRegionInfo[] regions) throws IOException {
// When AC is used, it should be configured as the 1st CP.
// In Master, the table operations like create, are handled by a Thread pool but the max size
// for this pool is 1. So if multiple CPs create tables on startup, these creations will happen
// sequentially only.
// Related code in HMaster#startServiceThreads
// {code}
// // We depend on there being only one instance of this executor running
// // at a time. To do concurrency, would need fencing of enable/disable of
// // tables.
// this.service.startExecutorService(ExecutorType.MASTER_TABLE_OPERATIONS, 1);
// {code}
// In future if we change this pool to have more threads, then there is a chance for thread,
// creating acl table, getting delayed and by that time another table creation got over and
// this hook is getting called. In such a case, we will need a wait logic here which will
// wait till the acl table is created.
if (AccessControlLists.isAclTable(desc)) {
this.aclTabAvailable = true;
} else if (!(TableName.NAMESPACE_TABLE_NAME.equals(desc.getTableName()))) {
if (!aclTabAvailable) {
LOG.warn("Not adding owner permission for table " + desc.getTableName() + ". "
+ AccessControlLists.ACL_TABLE_NAME + " is not yet created. "
+ getClass().getSimpleName() + " should be configured as the first Coprocessor");
} else {
String owner = desc.getOwnerString();
// default the table owner to current user, if not specified.
if (owner == null)
owner = getActiveUser().getShortName();
final UserPermission userperm = new UserPermission(Bytes.toBytes(owner),
desc.getTableName(), null, Action.values());
// switch to the real hbase master user for doing the RPC on the ACL table
User.runAsLoginUser(new PrivilegedExceptionAction<Void>() {
@Override
public Void run() throws Exception {
AccessControlLists.addUserPermission(c.getEnvironment().getConfiguration(),
userperm);
return null;
}
});
}
}
}