hdfs选择datanode策略

前几天报错如下: Error: org.apache.hadoop.ipc.RemoteException(java.io.IOException): File /user/xxx/part-r-00002 could only be replicated to 0 nodes instead of minReplication (=1). There are 11 datanode(s) running and no node(s) are excluded in this operation. at org.apache.hadoop.hdfs.server.blockmanagement.BlockManager.chooseTarget(BlockManager.java:1327) at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.getAdditionalBlock(FSNamesystem.java:2278) at org.apache.hadoop.hdfs.server.namenode.NameNodeRpcServer.addBlock(NameNodeRpcServer.java:480)……

JS Flowchart Diagrams

Usage

st=>start: Start|past:>http://www.google.com[blank]
e=>end: End:>http://www.google.com
op1=>operation: My Operation|past
op2=>operation: Stuff|current
sub1=>subroutine: My Subroutine|invalid
cond=>condition: Yes
or No?|approved:>http://www.google.com
c2=>condition: Good idea|rejected
io=>inputoutput: catch something...|request

st->op1(right)->cond
cond(yes, right)->c2
cond(no)->sub1(left)->op1
c2(yes)->io->e
c2(no)->op2->e

JS Sequence Diagrams

Usage

Andrew->China: Says Hello
Note right of China: China thinks\nabout it
China-->Andrew: How are you?
Andrew->>China: I am good thanks!

从日志分析impala查询慢的原因(2)

上一篇文章看到 requestTblLoadAndWait 速度缓慢,继续分析。先上日志 # impalad日志 I0103 15:09:35.503356 27319 impala-beeswax-server.cc:170] query(): query=select distinct(disp_id) from default_impala.kpi_disp_user_info_stat_tb where……