"Hadoop" 6, Hadoop installation error handling

Error 1, time cannot be synchronized

2014.12.18
To do the synchronization time, perform the command operation:
[Email protected] ~]#/usr/sbin/ntpdate pool.ntp.org
Will error:
Name server cannot be used, Exiting18 Dec 19:39:39 ntpdate[3592]: Name server cannot is used, reason:temporary failure in Name resolution

1, first confirm that the NTPD service is open.
[[email protected] ~]#/etc/init.d/ntpd start
Starting ntpd: [OK]

2, check whether your DNS is added, if not added, you use the add up.
To edit a file:

[Email protected] ~]# vi/etc/resolv.conf

Generated by NetworkManager

NameServer 202.99.144.12
NameServer 202.99.143.11

or add a generic dns:nameserver 8.8.8.8

Test:
[Email protected] ~]# ntpdate pool.ntp.org
Dec 13:40:32 ntpdate[23722]: The NTP socket is on use, exiting
[Email protected] ~]# date-r
Thu, Dec 2014 13:40:36 +0800

http://blog.itpub.net/27099995/viewspace-1372167/

2, when the Cloudera-manager-installer.bin installation is not complete
That is, when the installation of the time is not connected to the wrong time, we directly behind the corresponding USR/SHARE/CMF folder all deleted is so hungry

3, one problem is that Yum source cannot install Cloud-manager

After the Yum source update needs to import the key value, otherwise the error is as follows, cannot install the related package.
Total size:42 M
Downloading Packages:
Warning:rpmts_HdrFromFdno:Header V3 rsa/sha1 Signature, key ID C105b9de:nokey
Retrieving Key from File:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-5

GPG Key retrieval failed: [Errno] Could not open/read file:///etc/pki/rpm-gpg/RPM-GPG-KEY-CentOS-5

Workaround:
1. Install the following toolkit first:
RPM-UHV http://dag.wieers.com/packages/rpmforge-release/rpmforge-release-0.3.4-1.el4.rf.i386.rpm
[Email protected] yum.repos.d]# rpm–import/etc/pki/rpm-gpg/rpm-gpg-key-6

Generally, after the installation is complete, install this can be done.
If not, check the system's
Edit Yum source file: Replace rpm-gpg-key-centos-5 with the corresponding KEY file name: rpm-gpg-key-centos-6 (replace All)
[Email protected] yum.repos.d]# VI Centos-base-sohu.repo
[Base]
name=centos-$releasever-base-sohu.com
Mirrorlist=http://mirrorlist.centos.org/?release=releasever&arch=Basearch&repo=os
baseurl=http://mirrors.sohu.com/centos/ ReLeaseveR/os/ basearch/
Gpgcheck=1
Gpgkey=file:///etc/pki/rpm-gpg/rpm-gpg-key-centos-6

4, Error:rpmts_HdrFromFdno:headerRead Failed:hdr blob (982592): Bad, read returned 32296

The main reason for a lot of mistakes here is that the RPM package is broken, and I'll just download it all.

5, CDH installation failed, unable to authenticate

Perhaps we have previously configured the root user to do a secret key, we want to all users to the secret key!!!!

6, installation error cannot open "/var/log/cloudera-scm-agent//cloudera-scm-agent.log" read data: There is no file or directory

This error is because the Ip/hostname in the/etc/hosts is not correct, and then select Retry after the repair. This error is because the Ip/hostname in the/etc/hosts is not correct, and then select Retry after the repair. Cloudera Manager will first rollback all previous operations and revert to the original state before reloading. Cloudera Manager will first rollback all previous operations and revert to the original state before reloading.

7, installation failure. The detection signal from the Agent cannot be received.

请确保主机的名称已正确配置。请确保端口 7182 可在 Cloudera Manager Server 上访问（检查防火墙规则）。请确保正在添加的主机上的端口 9000 和 9001 空闲。检查正在添加的主机上 /var/log/cloudera-scm-agent/ 中的代理日志（某些日志可在安装详细信息中找到）。

Could not find config file/var/run/cloudera-scm-agent/supervisor/supervisord.conf

The solution to this error is:

After we have modified our/etc/hosts file, we have to restart the service cloudera-scm-agent

Service Cloudera-scm-agent Restart

8. Cannot be displayed after installing cm

9, 7180 interface cannot open

The experiment with Impala met Cloudera-scm-server dead but PID file exists problem.
This is how the problem occurs:
First, I installed the Cloudera Manager on 5 virtual machines configured for 2 cores of 4GB RAM, and then installed CDH and Impala with it.
And then everything is normal, and I'm going to import 100GB of data into the cluster's HDFs.
And then everything is still normal, but suddenly one day, Cloudera Manager's web interface is inaccessible (the 7180 port of that), Hadoop can also be used, Impala hangs off
Before the virtual machine with large memory installed hundreds of nodes of Impala are no problem, it is estimated that the memory is not enough (think of a single machine 2GB memory before the cluster, loaded soon Cloudera Manager will also not open, at that time because of the use of this cluster will be more than think of each), Looking at the primary node (that is, the node running Cloudera Manager, Cloudera-scm-server), there is only 400MB left in the memory.
It appears that the installation of Impala should be at least 8GB of stand-alone memory configuration.

However, we tried to solve the problem:
First of all, explain the experimental environment:
Hardware configuration above has said that the operating system is CentOS6.4 64-bit, software cdh4.5.0.30+impala1.2.1.850
The procedure is as follows:

Copyright NOTICE: This article for Bo Master original article, without Bo Master permission not reproduced.

"Hadoop" 6, Hadoop installation error handling