hbase关于hlog的几个疑问

最近在做一些初步的性能调优，对于一部分数据丢失，是可以容忍的。

所以我打算对hbase做一个写数据不写hlog的设置操作。

不过现在0.98版本，put.setWriteToWAL(false);是马上就要淘汰的方法了。
现在新的方法是？

还有，我在HDFS上找报这个信息，后面我去master的log日志看，
报：
hbase目录下有OldWALs和WALs两个目录，这个log清理是清理哪个？ 两个都给清理了？
2015-06-07 23:07:05,165 DEBUG [FifoRpcScheduler.handler1-thread-4] util.FSTableDescriptors: Exception during readTableDecriptor. Current table name = _LOCAL_IDX_INTEGRATE
org.apache.hadoop.hbase.TableInfoMissingException: No table descriptor file under hdfs://masters/hbase/data/default/_LOCAL_IDX_INTEGRATE
        at org.apache.hadoop.hbase.util.FSTableDescriptors.getTableDescriptorFromFs(FSTableDescriptors.java:510)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.getTableDescriptorFromFs(FSTableDescriptors.java:487)
        at org.apache.hadoop.hbase.util.FSTableDescriptors.get(FSTableDescriptors.java:173)
        at org.apache.hadoop.hbase.master.HMaster.getTableDescriptors(HMaster.java:2821)
        at org.apache.hadoop.hbase.protobuf.generated.MasterProtos$MasterService$2.callBlockingMethod(MasterProtos.java:41447)
        at org.apache.hadoop.hbase.ipc.RpcServer.call(RpcServer.java:2093)
        at org.apache.hadoop.hbase.ipc.CallRunner.run(CallRunner.java:101)
        at org.apache.hadoop.hbase.ipc.FifoRpcScheduler$1.run(FifoRpcScheduler.java:74)
        at java.util.concurrent.Executors$RunnableAdapter.call(Executors.java:471)
        at java.util.concurrent.FutureTask.run(FutureTask.java:262)
        at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
        at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
        at java.lang.Thread.run(Thread.java:745)
2015-06-07 23:26:47,935 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn1%2C60020%2C1433128563438.1433726190963
2015-06-07 23:26:47,938 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn2%2C60020%2C1433128545540.1433729773837.meta
2015-06-07 23:26:47,938 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn3%2C60020%2C1433128563674.1433729790761
2015-06-08 00:26:47,933 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn1%2C60020%2C1433128563438.1433729791187
2015-06-08 00:26:47,934 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn1%2C60020%2C1433128563438.1433733391546
2015-06-08 00:26:47,934 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn2%2C60020%2C1433128545540.1433729773830
2015-06-08 00:26:47,934 DEBUG [master:nn1:60000.oldLogCleaner] master.ReplicationLogCleaner: Didn't find this log in ZK, deleting: dn2%2C60020%2C1433128545540.1433733374120.meta

都可以删除掉。
应该是public void setDurability(Durability d)


楼主可以下载这个删除掉
hadoop2.2.0帮助手册下载API及HBase 0.98.1-hadoop2 API

yuwenge 发表于 2015-6-8 16:44
都可以删除掉。
应该是public void setDurability(Durability d)

看到了，谢谢回复~

ASYNC_WAL Write the Mutation to the WAL asynchronously

FSYNC_WAL Write the Mutation to the WAL synchronously and force the entries to disk.

SKIP_WAL Do not write the Mutation to the WAL

SYNC_WAL Write the Mutation to the WAL synchronously.

USE_DEFAULT If this is for tables durability, use HBase's global default value (SYNC_WAL).