impala常见错误记录
Posted 缪
tags:
篇首语:本文由小常识网(cha138.com)小编为大家整理,主要介绍了impala常见错误记录相关的知识,希望对你有一定的参考价值。
java.net.ConnectException: Call From YZJHZ1H007640P/10.174.110.12 to impala0:12900 failed on connection exception: java.net.ConnectException: Connection refused;
修改core-site.xml 改为9000端口
E0531 21:04:29.315419 31501 impala-server.cc:243] ERROR: short-circuit local reads is disabled because
- dfs.client.read.shortcircuit is not enabled.
配置hdfs.xml,加上该选项
E0531 21:24:52.664230 43354 impala-server.cc:243] ERROR: block location tracking is not properly enabled because
- dfs.client.file-block-storage-locations.timeout.millis is too low. It should be at least 10 seconds.
dfs.client.file-block-storage-locations.timeout.millis 选项设为10000 ,默认1000(1s)
java.net.ConnectException: connect(2) error: Not a directory when trying to connect to ‘/var/lib/hadoop-hdfs/dn_socket/‘
java.io.IOException: bad socket path ‘/var/lib/hadoop-hdfs/dn_socket/‘. The socket path must not end in a slash.
配置路径没有斜线
catalogd
E0602 17:14:12.026329 35744 catalog.cc:70] java.lang.NoClassDefFoundError: org/apache/hadoop/conf/Configuration
缺少各种库
yum remove 把openjdk相关卸载掉
yum remove java-1.8.0-openjdk
yum remove java-1.8.0-openjdk-headless
无法创建数据库 提示/user/hive/warehouse禁止写入
hadoop fs -chmod 777 /user/hive/warehouse
无法创建表,提示索引大于767字节
修改库的编码位latin1
alter database impala set latin1;
impala查询慢
hadoop datanode 只能启动一个,原因是VERSION文件同步了,导致只能启动一个datanode,避免的方法是不要同步hadoopdata文件夹,单独存放程序和数据文件
hadoop断电重启以后出现的问题
误操作,同步的文件切记不能同步hadoopdata目录
删除表的时候出现SQL语法错误异常
1、更新JDBC版本
2、修改hive配置,添加或修改
<property>
<name>hive.metastore.uris</name>
<value>thrift://192.168.1.188:9083</value>
</property>
启动hive
nohup hive --service metastore >/dev/null 2>&1 &
以上是关于impala常见错误记录的主要内容,如果未能解决你的问题,请参考以下文章