Installing Almond

There are currently packages for Ubuntu and RedHat to install Almond, which you can download from github. You should preferably put them in your repository of choice, since the installation have dependencies to a number of packages and libraries.

You can of course install the service without putting it in a repository, for instance on Ubuntu you type the following (as root) to install the application:


    dpkg -i almond-monitor_0.9.18.deb


    apt-get -f install

After installing the packages you most likely want to stop it:


    systemctl stop almond.service


    systemctl stop howru.service

and make sure you have the correct plugin settings in plugins.conf.

After stopping the service, check the configuration file for changes and edit the plugins file for your needs. This file os shortly described here.
Finally start the services again and enjoy your API.


    systemctl start almond.service


    systemctl start howru.service

Installing from source

If you are running a Linux distribution that does support the prebuilt packages you can install Almond from source.

PREREQUISITES

In order to run the install script you need to have a c compiler, make and automake installed.

gcc
make
automake

Make sure you have the dependecies for the source installation, meaning headers for:

json-c
ssl
rdkafka

For the default Nagios plugins to work you also need to have the following installed:

ksh
sysstat
python3
perl
ython3-psutil (pip3 install psutil)
ping
ps

RUN THE INSTALL SCRIPT

Run the script ./install_almond.sh
This script will compile and install Almond monitoring daemon and copy initial configuration and plugin files.

COMPILE WITH AUTOMAKE

If you for some reason do not want to use the install_almond.sh-script to compile and install Almond, you can also compile the source code yourself with automake, with the familiar configure/make/make install commands.

Configure can be run in three ways:

./configure --prefix=/opt/almond	standard compile of Almond with Kafka support
./configure --prefix=/opt/almond --enable-avro	compile Almond with avro support for Kafka
./configure --prefix=/opt/almond --disable-kafka	compiles a minimal Almond without Kafka support

make
make install

Note If you compile with avro support for the Almond Kafka producer also make sure to register the schema used by the Almond Kafka producer.
It should look something like this:


        curl -X POST http://localhost:18081/subjects/almond-monitor-topic-value/versions \ 

        -H "Content-Type: application/vnd.schemaregistry.v1+json" \ 

        -d @<(echo '{"schema": '"$(jq -Rs . < plugins_status.avsc)"'}')

where you of course must supply the correct url and path to your Kafka registry server.

Once installed (either through automake or install_almond.sh) you can start Almond with the following command:


        /opt/almond/start_almond.sh

Note Compiling Almond will not install HowRU, which is a Python application.