README¶
Deploy a single node InfluxDB with Ansible.
Summary¶
This repo builds a generic InfluxDB node based on the dynamic inventory generated by https://gitlab.com/ska-telescope/sdi/heat-cluster, which in turn relies on ansible collections from https://gitlab.com/ska-telescope/sdi/systems-common-roles
Checkout deploy-influxdb (this repo) and pull the dependent collections with:
git clone git@gitlab.com:ska-telescope/sdi/deploy-influxdb.git
cd deploy-influxdb
make install
Ansible vars for the cluster¶
In order to define the architecture of the cluster, one needs to describe, at a minimum, the dbnode_influxdb.
In influxdb_cluster_vars.yml, it is described a cluster containing 1 dbnode_influxdb (there is always just 1).
Molecule testing¶
Before running the build stage to deploy the InfluxDB node on the Openstack environment we encourage testing the ansible collection roles by using molecule.
The first step is to install the required python packages (we strongly recommend that a virtual environment is set). This can be acomplished with:
virtualenv venv && . venv/bin/activate
pip3 install -r requirements.txt
Molecule tests can then be run by doing:
make test
These tests do not require access to an openstack platform, they are fully local, but they need a lot of resources, at least 8 Gb of free available memory are recommended. The tests are also time consuming taking several minutes to complete.
For debugging purposes one can run the molecule tests without destroying the test container:
make molecule
With this option, resuming the tests after fixing any bugs that are found will be faster, but occasionally one may need to destroy the test container by doing:
make destroy
Setting your OpenStack environment¶
You need to use an adequate Openstack keypair name. This maybe be your usual personal keypair (which is most likely also your username) or a keypair defined just for this purpose. You also need your own RC file for the Openstack platform that you will be using. This file is available in shell or yaml formats. You should be able to retrieve it from the dashboards on the Openstack platform GUI.
If you choose the Openstack RC shell file, you should then source it (your password may be asked) and test your connection to the Openstack API:
. openrc.sh
openstack server list
Alternatively, if you prefer the yaml file (cloud.yaml) you should place it in the ~/.config/openstack/ folder. You may need to add your password to the auth section in that file. Do not forget to test the connection.
Referencing custom configuration¶
In order not to have to specify the CLUSTER_KEYPAIR, PRIVATE_VARS and
INVENTORY_FILE variables every time a make command is issued, we recommend that you create a
PrivateRules.mak file in the deploy-influxdb root directory specifying these
three variables (these need to be changed accordingly to each specific
purpose):
CLUSTER_KEYPAIR=your-openstack-key-name
PRIVATE_VARS=./influxdb_cluster_vars.yml
INVENTORY_FILE=./inventory_dev_cluster
Running the build¶
The build process is broken into two phases:
create the VM and do the common install steps
build influxdb
This can be run simply with:
make build
Or, in case you didn’t define the PrivateRules.mak with:
make build CLUSTER_KEYPAIR=your-openstack-key-name PRIVATE_VARS=./influxdb_cluster_vars.yml INVENTORY_FILE=./inventory_dev_cluster
You can also break the make build command into the two steps comprising it:
# create the VMs and do the common install
make build_nodes
# build InfluxDB
make build_influxdb
Destroying the nodes¶
Caution: The following command deletes everything that was installed and destroys the cluster specified by the PRIVATE_VARS variable.
In case you want to delete the whole stack issue the make clean command.
Help¶
Run make to get the help:
make targets:
mbuild Build nodes and influxdb
mbuild_common apply the common roles
mbuild_docker apply the docker roles
mbuild_influxdb Build influxdb
mbuild_nodes Build nodes based on heat-cluster
mbuild_vms Build vms only based on heat-cluster
mcheck_nodes Check nodes based on heat-cluster
mclean destroy the nodes - CAUTION THIS DELETES EVERYTHING!!!
mclean_nodes destroy the nodes - CAUTION THIS DELETES EVERYTHING!!!
mdestroy run molecule destroy - cleanup after make molecule
mhelp show this help.
minstall Install dependent ansible collections
mlint Lint check playbook
mmolecule run molecule tests but don't destroy container
mreinstall reinstall collections
mrtest run make $(STAGE) using gitlab-runner - default: test
mtest run molecule tests locally outside gitlab-runner
mvars Variables
mverify rerun molecule verify - must run make molecule first
make vars (+defaults):
mANSIBLE_USER ubuntu## ansible user for the playbooks
mCI_APPLICATION_TAG $(shell git rev-parse --verify --short=8 HEAD)
mCI_BUILD_TOKEN ""
mCI_ENVIRONMENT_SLUG development
mCI_JOB_ID job$(shell tr -c -d '0123456789abcdefghijklmnopqrstuvwxyz' </dev/urandom | dd bs=4 count=1 2>/dev/null;echo)
mCI_PIPELINE_ID pipeline$(shell tr -c -d '0123456789abcdefghijklmnopqrstuvwxyz' </dev/urandom | dd bs=8 count=1 2>/dev/null;echo)
mCLUSTER_KEYPAIR piers-engage## key pair available on openstack to be put in the created VMs
mCOLLECTIONS_PATHS ./collections
mCONTAINERD true
mDEBUG false
mDOCKERFILE Dockerfile
mDOCKER_VOLUMES /var/run/docker.sock:/var/run/docker.sock
mEXECUTOR docker
mEXTRA_VARS ?=
mGITLAB_USER ""
mGITLAB_USER_EMAIL "nobody@example.com"
mIGNORE_NVIDIA_FAIL false
mINVENTORY_FILE ./inventory_influxdb_cluster##inventory file to be generated
mNVIDIA false
mPRIVATE_VARS ./influxdb_cluster_vars.yml##template variable for heat-cluster
mREGISTRY_TOKEN ""
mREPOSITORY_TOKEN ""
mSTAGE test## Molecule stage name
mTAGS ?=
mV ?=