CDH problems encountered when deploying a cluster

CDH: Full name Cloudera ' s distribution including Apache Hadoop
CDH version-derived Hadoop is an open source project, so many companies are commercializing this foundation, and Cloudera has made a corresponding change to Hadoop.
Cloudera Company's release, we call this version CDH (Cloudera distribution Hadoop). 　　So far, there are 5 versions of CDH, of which the first two are no longer updated, and the last two, respectively, are CDH4, which evolved on the basis of Apache Hadoop 2.0.0, CDH5, which are updated every once in a while.　　Recently installed CDH5 installation encountered a ton of problems, I tidied up for the first installation of everyone, hoping to save everyone's working time and study time. I am using the CentOS system offline installation method, installed a lot of documents online, much the same: Ubuntu installation case http://www.aboutyun.com/thread-9107-1-1.html CentOS installation case http:/ Www.aboutyun.com/thread-10852-1-1.html CDH Working with Documents http://www.cloudera.com/content/www/zh-CN/documentation/　　Enterprise/5-3-x/topics/introduction.html You can refer to the above document for reference. Question 1:initializing on the master node CM5the databaseError: ld-linux.so.2 bad ELF interpreter fix: Install GLIBC and glibc.i686 Issue 2: Error: ProtocolError: <protocolerror for 127.0.0.1/RPC2: 401 unauthorized> 解决：$> ps -ef | grep supervisord$> kill-9 <processid>/opt/cm-5.6.0/etc/init.d/cloudera-scm-agent Restart issue 3:server startup, the log prompt port is occupied. WORKAROUND: Delete the Java process. Problem 4:web installation, current managed host display is local address resolved: Comment/etc/hosts loaclhost, check agent log for error. Restart all agents Restart Server issue 4:web database settings, login denied: Grant all privileges on * * to ' hive ' @ ' cdh1 ' identified by ' 123456 ' with Grant Opti On; Specify host name or IP address problem 5:web installation, the cluster settings HDFs format fails to resolve: Remove the original/DFS problem 6:web installation, the cluster settings HDFs Create/tmp failed to resolve: NTP must start the server, not light with the command synchronization. (This does not seem to be the problem, but the NTP service must be started) also appears, try again. Issue 7:web Management page prompt for time skew resolution: Check ntpdc-c Loopinfoname or service not knownvim/etc/hosts add native IP corresponding to localhost address

CDH problems encountered when deploying a cluster