CDH升级导致EsgynDB启不来的问题分析

最新推荐文章于 2022-12-07 11:03:39 发布

数据源的港湾

最新推荐文章于 2022-12-07 11:03:39 发布

阅读量262

点赞数 1

分类专栏： QianBase/EsgynDB/Trafodion QianBase/EsgynDB诊断

本文链接：https://blog.csdn.net/Post_Yuan/article/details/90114765

版权

QianBase/EsgynDB/Trafodion 同时被 2 个专栏收录

263 篇文章 3 订阅

订阅专栏

QianBase/EsgynDB诊断

65 篇文章 0 订阅

订阅专栏

现象

近来有一个客户那个经常性的EsgynDB起不来。从CDH Manager看起来Hadoop集群是正常的，但trafodion用户执行hbcheck确显示HBase状态异常。

原因

使用trafodion用户执行hbcheck显示Unavailable，但CDH Manager中显示正常。于是执行hbase shell想检查是否正常执行相关HBase命令，发现hbase可执行文件。
检查HBase和hadoop可执行文件，发现/usr/bin/hbase和/usr/bin/hadoop等相关的软链接被ls到 /opt/cloudera/parcels/CDH-5.6.0-1.cdh5.6.0.p0.45目录下，但目录并不存在，原因是此环境中的CDH做了升级，从原来的CDH 5.6升级到CDH 5.7，导致软链接问题。

ll /usr/bin/hbase
ll /usr/bin/hadoop
[root@cdh1 ~]# ll /etc/alternatives/hbase
lrwxrwxrwx 1 root root 59 May  9 09:40 /etc/alternatives/hadoop -> /opt/cloudera/parcels/CDH-5.6.0-1.cdh5.6.0.p0.45/bin/hbase
[root@cdh1 ~]# ll /etc/alternatives/hadoop
lrwxrwxrwx 1 root root 59 May  9 09:40 /etc/alternatives/hadoop -> /opt/cloudera/parcels/CDH-5.6.0-1.cdh5.6.0.p0.45/bin/hadoop

解决

修改相关软链接

rm -rf /etc/alternatives/hbase
rm -rf /usr/bin/hadoop
rm -rf  /etc/alternatives/zookeeper-client
rm -rf /etc/alternatives/zookeeper-server
rm -rf  /etc/alternatives/zookeeper-server-cleanup
rm -rf /etc/alternatives/zookeeper-server-initialize
ln -s /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/hbase /etc/alternatives/hbase
ln -s /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/hadoop /usr/bin/hadoop
ln -s  /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/zookeeper-client  /etc/alternatives/zookeeper-client
ln -s  /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/etc/zookeeper/conf.dist  /etc/alternatives/zookeeper-conf
ln -s  /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/zookeeper-server  /etc/alternatives/zookeeper-server
ln -s  /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/zookeeper-server-cleanup  /etc/alternatives/zookeeper-server-cleanup
ln -s  /opt/cloudera/parcels/CDH-5.7.6-1.cdh5.7.6.p0.6/bin/zookeeper-server-initialize  /etc/alternatives/zookeeper-server-initialize

使用trafodion用户重新sqgen生成环境变量

pdsh $MY_NODES "cds; mv sqconfig.db sqconfig.db.bak"
sqgen

执行hbcheck重新检查

hbcheck
[trafodion@cdh1 ~]$ hbcheck
Stderr being written to the file: /home/trafodion/esgynDB_server-2.4.1/logs/hbcheck.log
ZooKeeper Quorum: cdh3.cluster.cm,cdh2.cluster.cm,cdh1.cluster.cm, ZooKeeper Port  : 2181

HBase is available!

HBase version: 1.2.0-cdh5.7.6
HMaster: cdh1.cluster.cm,60000,1557366248622

Number of RegionServers available:3
RegionServer #1: cdh1.cluster.cm,60020,1557366248847
RegionServer #2: cdh2.cluster.cm,60020,1557367662923
RegionServer #3: cdh3.cluster.cm,60020,1557366248416

Number of Dead RegionServers:0
Number of regions: 2429
Number of regions in transition: 0
Average load: 809.6666666666666