Ambari install script location(s) - hadoop

I'm setting up a HDP 2.1 cluster with Apache Ambari. All servers run SLES 11 SP3. The setup fails if I select to install Ganglia because of some dependencies:
Installing package apache2?mod_php* ('/usr/bin/zypper --quiet install --auto-agree-with-licenses --no-confirm apache2?mod_php*')
Problem: apache2-mod_php53-5.3.17-0.27.1.x86_64 conflicts with apache2-mod_php5 provided by apache2-mod_php5-5.2.14-0.7.30.50.1.x86_64
Solution 1: Following actions will be done:
do not install apache2-mod_php5-5.2.14-0.7.30.50.1.x86_64
deinstallation of php5-5.2.14-0.7.30.50.1.x86_64
deinstallation of php5-xmlwriter-5.2.14-0.7.30.50.1.x86_64
[... more PHP 5.2.x packages ...]
Solution 2: do not install apache2-mod_php53-5.3.17-0.27.1.x86_64
Apparently the Regex picks the 5.3 version, a 5.2 version would be available though.
So my question is: Where is the install script stored, that Ambari is running here? I would like to replace the regex with the correct version of the package.

Information about what packages are to be installed is stored in
/var/lib/ambari-server/resources/stacks/HDP/2.0.6/services/GANGLIA/metainfo.xml
Change the value and restart the Ambari Server for the changes to take effect.

Related

Maven version error while trying to install ambari on centos

I am trying to install ambari 2.7.5 on centos 7. I am following the instructions given on this link. When I run the build command for maven I get the following error:
I have installed maven 3.6.3 . And I could not find maven 3.1.0 in apache downloads. So how to resolve this error?
#VK for Ambari-web you have to edit some files and set your versions. If you are having hard time with npm/node version make sure they match the posted versions I gave. In centos7 I am using the epel repository. Sounds like you already got current maven so good to go there.
Also this post is pretty much a duplicate of:
Building Ambari 2.7.5 on CentOS 7 from source, Worked 2 weeks ago, now fails
Its also discussed here:
Ambari 2.7.5 installation failure on CentOS 7
And a jira with another bower file sample:
https://issues.apache.org/jira/browse/AMBARI-25519

Installing specific version of hadoop-client on Ubuntu

I have installed a CDH5(cdh5.15.1 version) cluster on ubuntu nodes and I want to install hadoop-client on one ubuntu host (which is not managed by cloudera).
I added cloudera repo on my ubuntu host but when I run apt-get install hadoop-client , by default it install version 5.16. However I want to install version 5.15.1 version for hadoop-client( which is available in repo).
The problem is I am not able to figure out which version number should I give . The deb package name for hadoop client cdh5.15.1 is:
hadoop-client_2.6.0+cdh5.15.1+2822-1.cdh5.15.1.p0.4~xenial-cdh5.15.1_all.deb
I have to run apt-get install hadoop-client=< version number > so that it picks up and install the specific version.
Please help me to figure out the exact version number.
Thanks in adv. :)
use apt-cache policy hadoop-client to see versions available from your package CDH
EDIT :
use apt-cache madison hadoop-client instead to see all versions available in your packages.
You can also use apt list -a hadoop-client which has the same purprose

How to install Tachograph (a monitoring tool of MonetDB Jul2015 release) in Ubuntu 14.04

I have installed the latest Jul2015 release, and I would like to use the
latest tools tachograph to determine the progress of query execution.
According to https://www.monetdb.org/Documentation/Manuals/MonetDB/Profiler/tachograph,
Tachograph is available as of Jul2015 release
But I do not know how to install it. My system is ubuntu 14.04.
I have tried to use sudo apt-get install monetdb-tools but failed to
locate the package monetdb-tools.
Any suggestion would be helpful. Thanks.
Tachograph is in monetdb-client-tools. Also, see https://dev.monetdb.org/downloads/deb/ for instructions on how to add the MonetDB repository to your system.

How to install Logstash for NodeJs on Windows 7

I want to install Logstash for NodeJs on windows 7, but I am not able to find proper steps for the same.
Can any one please help!
There is the option of node-logstash if you want a node.js alternative to Logstash. This isn't something I'm using myself (I'm using nxlog in Windows instead) but it looks like a decent alternative to the standard JRuby Logstash if you need to forward logs from Windows.
Instructions from the readme are below:
Installation
Install NodeJS, version >= 0.10, or io.js.
Install build tools
Debian based system: apt-get install build-essential
Centos system: yum install gcc gcc-c++ make
Install zmq dev libraries: This is required to build the node zeromq module.
Debian based system: apt-get install libzmq1. Under recent releases, this package is present in default repositories. On ubuntu lucid, use this ppa. On debian squeeze, use backports.
Centos 6: yum install zeromq zeromq-devel. Before, you have to add the rpm zeromq repo : curl http://download.opensuse.org/repositories/home:/fengshuo:/zeromq/CentOS_CentOS-6/home:fengshuo:zeromq.repo > /etc/yum.repos.d/zeromq.repo
Clone repository: git clone git://github.com/bpaquet/node-logstash.git && cd node-logstash
Install dependencies: npm install.
The executable is in bin/node-logstash-agent
You have scripts in dists folder to build packages. Actually, only debian is supported.
As per the comment, logstash has nothing to do with nodejs.
What you're looking to do is install Logstash on Windows, something that you can find out about by using google, there will be loads of guides out there describing how to do this.
You would then need to configure logstash to look in the right location for the log files it needs to process, and then set up filters to handle nodejs style logs (which as far as I understand aren't very well standardised). You then need to configure an output (logstash is essentially a unix pipe on steroids and needs somewhere to save the logs it has processed). Elasticsearch is the most common thing to save logs to.
Personally, in my environment, I would install logstash on a CentOS server, as it's a well established process, and ship the logs from your Windows 7 machine to the logstash server using either logstash forwarder or nxlog. That way you can have logs coming in from a number of different sources and you can still reboot your Windows machine every few days as required by Windows update without your logstash server going down.

how to determine the cloudera minor release in the one click install debian package ? (i.e., 5.1 ? 5.2 ?)

I've managed to get cloudera single node hadoop cluster up and running from this package: http://archive.cloudera.com/cdh5/one-click-install/precise/amd64/cdh5-repository_1.0_all.deb
my colleagues asked me what minor release of cloudera this is installing.. and i am stumped as to which. Is there some info, readme, config or license file that gives this information for cloudera hadoop distros once they are installed ?
Or maybe someone just knows which minor release the above URL will install (if you could provide that info along with a link to a documentation source that would be fantastic.)
thanks in advance
-chris
The one-click install repo currently points to the latest Cloudera version, which is 5.3.0 as of earlier this week.
To check the version you installed, just list the package name. There should be some version number like '5.2.x' appended to the package name. An example command:
dpkg -l | grep 'cloudera'

Resources