ClouderaManager安装踩过的坑（NTP/JDK/HTTP）

1、NTP时钟源同步问题

查看时钟源是否唯一且为master，如下图示有三个时钟源

ntpq -p

以下步骤将系统及硬件时钟同步至master

systemctl enable ntpd.service
ntpd
ntpdate -u 10.47.243.43
hwclock –systohc
systemctl restart ntpd.service

2、JAVA/JDK1.8.0_11问题

告警如下

[root@master cloudera-scm-server]# tail -200f cloudera-scm-server.out
JAVA_HOME=/usr/local/java/jdk1.8.0_11 /usr/share/cmf/bin/cmf-server: line 129: /usr/local/java/jdk1.8.0_11/bin/java: Permission denied
/usr/share/cmf/bin/cmf-server: line 129: exec:/usr/local/java/jdk1.8.0_11/bin/java: cannot execute: Permission denied

ClouderaManager默认从/ usr / java /文件夹下读取JAVA变量，建议将jdk-8u11-linux-x64.tar.gz压缩包解压至此目录下，步骤如下：

cd /usr/java
rm -rf cd jdk1.7.0_67-cloudera
cd /usr/local/java/
cp jdk-8u11-linux-x64.tar.gz /usr/java
cd /usr/java
tar -zxvf jdk-8u11-linux-x64.tar.gz

vi /etc/profile（在文件中添加以下缩进行环境变量及路径）

echo “JAVA_HOME=/usr/java/jdk1.8.0_11” >> /etc/environment（添加全局变量）
vi /etc/default/cloudera-scm-server（在该文件下添加以下语句）
export JAVA_HOME=”/usr/java/jdk1.8.0_11”

3、httpd本地parcel存储库网页无法打开的问题(文件及文件夹权限 chmod 777)
告警如下

[root@master httpd]# cat error_log [Sun Aug 26 03:39:01.287071 2018]
[auth_digest:notice] [pid 1326] AH01757: generating secret for digest
authentication … [Sun Aug 26 03:39:01.287656 2018]
[lbmethod_heartbeat:notice] [pid 1326] AH02282: No slotmem from
mod_heartmonitor [Sun Aug 26 03:39:01.289606 2018]
[mpm_prefork:notice] [pid 1326] AH00163: Apache/2.4.6 (CentOS)
OpenSSL/1.0.1e-fips mod_fcgid/2.3.9 configured – resuming normal
operations [Sun Aug 26 03:39:01.289623 2018] [core:notice] [pid 1326]
AH00094: Command line: ‘/usr/sbin/httpd -D FOREGROUND’ [Mon Aug 27
09:49:48.471984 2018] [core:error] [pid 24627] (13)Permission denied:
[client 10.210.17.129:59260] AH00035: access to
/yum.repos.ld/index.html denied (filesystem path
‘/var/www/html/yum.repos.ld/index.html’) because search permissions
are missing on a component of the path

根据日志提示无权限访问，将/yum.repos.ld/这个文件夹开放权限为777后，本地parcel存储库网页正常显示
4、格式化namenode节点失败(CM页面格式化)
ClouderaManager HDFS启动失败：提示格式化失败namenode format failed
clouderamanager启动界面告警如下

18/08/27 17:31:34 INFO namenode.FSNamesystem: ACLs enabled? false
18/08/27 17:31:34 INFO namenode.FSNamesystem: XAttrs enabled? true
18/08/27 17:31:34 INFO namenode.FSNamesystem: Maximum size of an
xattr: 16384 Running in non-interactive mode, and data appears to
exist in Storage Directory /data/dfs/nn. Not formatting.

hdfs日志告警如下

2018-08-28
14:49:37,350 INFO org.apache.hadoop.hdfs.server.common.Storage: Lock
on /data/dfs/nn/in_use.lock acquired by nodename 1544@master
2018-08-28 14:49:37,353 INFO
org.apache.hadoop.hdfs.server.common.Storage: Lock on
/tivoli/dfs/nn/in_use.lock acquired by nodename 1544@master 2018-08-28
14:49:37,354 WARN org.apache.hadoop.hdfs.server.namenode.FSNamesystem:
Encountered exception loading fsimage java.io.IOException: NameNode is
not formatted.
at org.apache.hadoop.hdfs.server.namenode.FSImage.recoverTransitionRead(FSImage.java:232)
at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFSImage(FSNamesystem.java:1150)
at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFromDisk(FSNamesystem.java:797)
at org.apache.hadoop.hdfs.server.namenode.NameNode.loadNamesystem(NameNode.java:614)
at org.apache.hadoop.hdfs.server.namenode.NameNode.initialize(NameNode.java:676)
at org.apache.hadoop.hdfs.server.namenode.NameNode.(NameNode.java:844)
at org.apache.hadoop.hdfs.server.namenode.NameNode.(NameNode.java:823)
at org.apache.hadoop.hdfs.server.namenode.NameNode.createNameNode(NameNode.java:1547)
at org.apache.hadoop.hdfs.server.namenode.NameNode.main(NameNode.java:1615)
2018-08-2814:49:37,363INFOorg.mortbay.log:Stopped
HttpServer2$SelectChannelConnectorWithSafeStartup@master:50070
2018-08-28 14:49:37,363 INFO
org.apache.hadoop.metrics2.impl.MetricsSystemImpl: Stopping NameNode
metrics system… 2018-08-28 14:49:37,364 INFO
org.apache.hadoop.metrics2.impl.MetricsSystemImpl: NameNode metrics
system stopped. 2018-08-28 14:49:37,364 INFO
org.apache.hadoop.metrics2.impl.MetricsSystemImpl: NameNode metrics
system shutdown complete. 2018-08-28 14:49:37,364 ERROR
org.apache.hadoop.hdfs.server.namenode.NameNode: Failed to start
namenode. java.io.IOException: NameNode is not formatted.
at org.apache.hadoop.hdfs.server.namenode.FSImage.recoverTransitionRead(FSImage.java:232)
at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFSImage(FSNamesystem.java:1150)
at org.apache.hadoop.hdfs.server.namenode.FSNamesystem.loadFromDisk(FSNamesystem.java:797)
at org.apache.hadoop.hdfs.server.namenode.NameNode.loadNamesystem(NameNode.java:614)
at org.apache.hadoop.hdfs.server.namenode.NameNode.initialize(NameNode.java:676)
at org.apache.hadoop.hdfs.server.namenode.NameNode.(NameNode.java:844)
at org.apache.hadoop.hdfs.server.namenode.NameNode.(NameNode.java:823)
at org.apache.hadoop.hdfs.server.namenode.NameNode.createNameNode(NameNode.java:1547)
at org.apache.hadoop.hdfs.server.namenode.NameNode.main(NameNode.java:1615)
2018-08-28 14:49:37,365 INFO org.apache.hadoop.util.ExitUtil: Exiting
with status 1 2018-08-28 14:49:37,367 INFO
org.apache.hadoop.hdfs.server.namenode.NameNode: SHUTDOWN_MSG:

主节点添加以下

echo ‘hdfs ALL=(ALL) NOPASSWD: ALL’ >> /etc/sudoers

清空以下路径下的所有文件，然后一定要在客户端界面的HDFS组件界面进行格式化操作并启动：
主节点

rm -rf /data/dfs/nn/*
rm -rf /tivoli/dfs/nn/*

从节点

rm -rf /data/dfs/dn/*
rm -rf /tivoli/dfs/dn/*

5、SPARK2未找到哈希文件
告警如下

2018-08-29 15:02:23,990 WARN
ParcelUpdateService:com.cloudera.parcel.components.LocalParcelManagerImpl:
Error reading hash file:
SPARK2-2.2.0.cloudera1-1.cdh5.12.0.p0.142354-el7.parcel.sha
java.io.FileNotFoundException:
/opt/cloudera/parcel-repo/SPARK2-2.2.0.cloudera1-1.cdh5.12.0.p0.142354-el7.parcel.sha
(Permission denied) 2018-08-29 15:02:23,993 WARN
ParcelUpdateService:com.cloudera.parcel.components.LocalParcelManagerImpl:
Invalid hash format found in
SPARK2-2.2.0.cloudera1-1.cdh5.12.0.p0.142354-el7.parcel.sha

权限问题，更改以下文件权限为644，并重启cloudera-scm-server服务后才能找到哈希文件。
/opt/cloudera/parcel-repo/SPARK2-2.2.0.cloudera1-1.cdh5.12.0.p0.142354-el7.parcel.sha