问题描述
我完全按照docs.datastax.com/en/datastax_enterprise/install/installDEBdse.html概述的方式使用了apt存储库安装方法
I used the apt repository installation approach exactly as outlined at docs.datastax.com/en/datastax_enterprise/install/installDEBdse.html
键。我有Oracle Server JDK 8（截止到今天）。 Miniconda的Python 2.7（今天也可以重新安装），使用默认设置并允许PATH变量开头。
There were no issues with the key. I have Oracle Server JDK 8 (latest as of today). Python 2.7 from Miniconda (also fresh install today), using defaults and allowing for prepending of PATH variables.
在安装之后，存在有关未满足的依赖项的错误：
Following installation, there are errors regarding unmet dependencies:
dse-full : Depends: dse (=4.8.4-1) but it is not going to be installed
Depends: dse-hive (=4.8.4-1) but it is not going to be installed
Depends: dse-pig (=4.8.4-1) but it is not going to be installed
Depends: dse-demos (=4.8.4-1) but it is not going to be installed
E: Unable to correct problems, you have held broken packages.
此服务器已完全裸装，因此Ubuntu环境是干净的并已更新。
This server was completely nuked, so the Ubuntu environment was clean and updated.
我已经删除了sources.list.d文件夹，获得了一个新密钥（删除该文件夹也删除了与任何密钥对的关联），并发现了完全相同的错误。
I've deleted the sources.list.d folder, obtained a fresh key (deleting this folder also removed associations with any key pairs), and found the exact same error.
这里不知道如何纠正和继续。
No idea here on how to correct and proceed.
可能与之相关的其他信息是讨厌的，因为在我之前的安装尝试中，apt仓库的http和https均抛出错误，因此下载文件并在安装后进行安装可执行文件。没有引发任何错误。但是，在尝试安装最新的OpsCenter时，出现与丢失的文件和文件夹有关的错误。我什至无法卸载DSE文档，因为从未创建/ usr / share / dse文件夹。
Possibly pertinent additional info is the nuking was because in my previous attempt to install had the apt repository throw errors with both http and https, so downloaded file and installed after making it executable. No errors were thrown. However, when attempting to install the newest OpsCenter, was given errors pertaining to missing files and folders. I could not even uninstall per DSE documentation, as the /usr/share/dse folder was never created.
更多信息关于：错误和环境
MORE INFORMATION RE: ERRORS AND ENVIRONMENT
环境-
$ lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description: Ubuntu 14.04.3 LTS
Release: 14.04
Codename: trusty
$ java -version
java version "1.8.0_72"
Java(TM) SE Runtime Environment (build 1.8.0_72-b15)
Java HotSpot(TM) 64-Bit Server VM (build 25.72-b15, mixed mode)
$ python --version
Python 2.7.5
缓存内容-
$ apt-cache madison dse-full | grep 4.8
dse-full | 4.8.4-1 | https://debian.datastax.com/enterprise/ stable/main amd64 Packages
dse-full | 4.8.3-1 | https://debian.datastax.com/enterprise/ stable/main amd64 Packages
dse-full | 4.8.2-1 | https://debian.datastax.com/enterprise/ stable/main amd64 Packages
dse-full | 4.8.1-1 | https://debian.datastax.com/enterprise/ stable/main amd64 Packages
dse-full | 4.8.0-1 | https://debian.datastax.com/enterprise/ stable/main amd64 Packages
REMOVAL-
$ sudo apt-get purge "dse*" datastax-agent
Package 'datastax-agent' is not installed, so not removed
Package 'dse-demos' is not installed, so not removed
Package 'dse-full' is not installed, so not removed
Package 'dse-hive' is not installed, so not removed
Package 'dse-libcassandra' is not installed, so not removed
Package 'dse-libhadoop-native' is not installed, so not removed
Package 'dse-libhadoop' is not installed, so not removed
Package 'dse-libhive' is not installed, so not removed
Package 'dse-liblog4j' is not installed, so not removed
Package 'dse-libmahout' is not installed, so not removed
Package 'dse-libpig' is not installed, so not removed
Package 'dse-libsolr' is not installed, so not removed
Package 'dse-libspark' is not installed, so not removed
Package 'dse-libsqoop' is not installed, so not removed
Package 'dse-libtomcat' is not installed, so not removed
Package 'dse-pig' is not installed, so not removed
Package 'dse' is not installed, so not removed
新安装-
$ sudo apt-get install dse-full
Reading package lists... Done
Building dependency tree
Reading state information... Done
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:
The following packages have unmet dependencies:
dse-full : Depends: dse (= 4.8.4-1) but it is not going to be installed
Depends: dse-hive (= 4.8.4-1) but it is not going to be installed
Depends: dse-pig (= 4.8.4-1) but it is not going to be installed
Depends: dse-demos (= 4.8.4-1) but it is not going to be installed
E: Unable to correct problems, you have held broken packages.
推荐答案
我自己走了这条路。我强烈建议以下内容：
I went down this path myself. I highly recommend the following:
1）从Ubuntu 12节点重新开始。 （对我而言，我使用了Amazon EC2 ami-15cbe87f。）
1) Start over with an Ubuntu 12 node. (For me, I used Amazon EC2 ami-15cbe87f.)
2）一旦节点启动，请下载并运行最新的Datastax Enterprise安装程序。这将升级群集，并将hadoop放置在/ usr / share / dse / resources / hadoop中。 （对我来说，我使用了DataStaxEnterprise-4.8.4.2016011118-linux-x64-installer.run）
2) Once the node launches, download and run the latest Datastax Enterprise installer. This upgrades the cluster, and puts hadoop home at /usr/share/dse/resources/hadoop. (For me, I used DataStaxEnterprise-4.8.4.2016011118-linux-x64-installer.run)
2）在/etc/dse/cassandra/cassandra.yaml中：
-设置群集名称
-使用内部up地址172.xx.xx.xx作为种子节点，listen_address，broadcast_address，broadcast_rpc_address
-使用0.0.0.0作为rpc_address
-使用PropertyFileSnitch并编辑/etc/dse/cassandra/cassandra-topology.properties。添加内部和外部IP的条目。
-＃Cassandra节点IP =数据中心：机架
-52.xx.xx.xx =美国东部：1a
-172.xx.xx.xx =我们- east：1a
-＃未知节点的默认值
-默认= us-east：1a
2) In /etc/dse/cassandra/cassandra.yaml: - set cluster name - use the internal up address 172.xx.xx.xx for seed node, listen_address, broadcast_address, broadcast_rpc_address - use 0.0.0.0 for rpc_address - use PropertyFileSnitch and edit /etc/dse/cassandra/cassandra-topology.properties. Add entries for both the internal and external IPs. -- # Cassandra Node IP=Data Center:Rack -- 52.xx.xx.xx=us-east:1a -- 172.xx.xx.xx=us-east:1a -- # default for unknown nodes -- default=us-east:1a
3）复制/ etc / dse / hadoop / taskcontroller.cfg到/etc/hadoop/taskcontroller.cfg。仍然有一段代码没有使用HADOOP_CONF_DIR。
3) Copy /etc/dse/hadoop/taskcontroller.cfg to /etc/hadoop/taskcontroller.cfg. There is still a piece of code that is not using HADOOP_CONF_DIR.
4）将Dir更改为/ usr / share / dse / resources / hadoop / native / Linux-amd64- 64 / bin和chmod 4750任务控制器
4) Change Dir into /usr/share/dse/resources/hadoop/native/Linux-amd64-64/bin and chmod 4750 task-controller
5）是，4750。
6）不要忘记编辑/var/lib/datastax-agent/conf/address.yaml以包含：
-stomp_interface：172.xx.xx.xx
-local_interface：172.xx.xx.xx
-use_ssl：0
6) Don't forget to edit /var/lib/datastax-agent/conf/address.yaml to contain: - stomp_interface: 172.xx.xx.xx - local_interface: 172.xx.xx.xx - use_ssl: 0
7）在某些时候，您需要返回并设置ssl。
7) At some point, you will need to go back and set up ssl.
8）sudo服务dse启动并检查/var/log/cassandra/output.log和system.log中的错误
8) sudo service dse start and check for errors in /var/log/cassandra/output.log and system.log
9）修复后，您可以删除您的集群并通过进入/ var / lib / cassandra并在那里删除所有内容来重新开始。 （请注意！）然后，返回到步骤8。重复操作，直到没有错误。
9) After a fix, you can delete your cluster and start over by going into /var/lib/cassandra and deleting everything there. (Be careful!) Then, go back to step 8. Repeat until no errors.
10）使用这种方法可以使纯Cassandra节点正常工作。然后，使用这种方法来使火花/ Hadoop节点正常工作。然后，您可以将它们结合起来并扩展集群。
10) Use this approach to get a Cassandra-only node working cleanly. Then, use this approach to get a spark/hadoop node working cleanly. Then, you can combine them and grow your cluster.
这篇关于使用apt信息库安装时，Ubuntu 14.04 LTS上的DataStax Enterprise 4.8.4安装错误的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持！