This tutorial shows how to set up a fully working Hydra server using our charm, MicroK8s and Juju
Set things up
Bootstrap a microk8s controller using juju 3.4 and create a new Juju model:
$ juju add-model hydra
Added 'hydra' model on microk8s/localhost with credential 'microk8s' for user 'admin'
Watch the Hydra charm transform the way to deploy, configure, integrate, and manage Hydra on any Kubernetes cloud
Hydra requires a way to persist data, in the case of our charm we enforce the usage of a postgreSQL database
As mentioned, we need a persistent way to store Hydra data, we are going to be using the postgresql-k8s charm
$ juju deploy postgresql-k8s --channel 14/stable --trust postgresql
Deployed "postgresql" from charm-hub charm "postgresql-k8s", revision 193 in channel 14/stable on ubuntu@22.04/stable
Once that is done (no need to wait for it to be ready) we can proceed in deploying hydra and integrate the 2 charms
$ juju deploy hydra --channel 0.2/stable
Deployed "hydra" from charm-hub charm "hydra", revision 278 in channel 0.2/stable on ubuntu@22.04/stable
Integrate with PostgreSQL
$ juju integrate postgresql:database hydra
after some time we should be able to inspect that all has been successfully deployed and connected
$ juju status --relations
Model Controller Cloud/Region Version SLA Timestamp
hydra microk8s-localhost microk8s/localhost 3.4.2 unsupported 15:57:02+01:00
App Version Status Scale Charm Channel Rev Address Exposed Message
hydra v2.2.0 waiting 1 hydra 0.2/stable 278 10.152.183.141 no installing agent
postgresql 14.10 active 1 postgresql-k8s 14/stable 193 10.152.183.100 no Primary
Unit Workload Agent Address Ports Message
hydra/0* blocked idle 10.1.245.156 Missing required relation with ingress
postgresql/0* active idle 10.1.245.160 Primary
Integration provider Requirer Interface Type Message
hydra:hydra hydra:hydra hydra_peers peer
postgresql:database hydra:pg-database postgresql_client regular
postgresql:database-peers postgresql:database-peers postgresql_peers peer
postgresql:restart postgresql:restart rolling_op peer
postgresql:upgrade postgresql:upgrade upgrade peer
you will notice that hydra is in a blocked status now, reason is that we will need to expose it via an ingress to keep moving forwards with the deployment
Integrate with an Ingress
Hydra requires an ingress relation to be healthy, this is due to the fact that its APIs need to be publicly available to be a useful component
Traefik is the charm of choice to make this happen
$ juju deploy traefik-k8s --channel stable
Deployed "traefik-k8s" from charm-hub charm "traefik-k8s", revision 176 in channel latest/stable on ubuntu@20.04/stable
We can now proceed with the integration to expose Hydra public APIs via the public-ingress relation
$ juju integrate hydra:public-ingress traefik-k8s
after some time we should be able to inspect that all has been successfully deployed and connected
$ juju status --relations
Model Controller Cloud/Region Version SLA Timestamp
hydra microk8s-localhost microk8s/localhost 3.4.2 unsupported 16:02:36+01:00
App Version Status Scale Charm Channel Rev Address Exposed Message
hydra v2.2.0 waiting 1 hydra 0.2/stable 278 10.152.183.141 no installing agent
postgresql 14.10 active 1 postgresql-k8s 14/stable 193 10.152.183.100 no Primary
traefik-k8s v2.11.0 active 1 traefik-k8s latest/stable 176 10.64.140.0 no
Unit Workload Agent Address Ports Message
hydra/0* active idle 10.1.245.156
postgresql/0* active idle 10.1.245.160 Primary
traefik-k8s/0* active idle 10.1.245.159
Integration provider Requirer Interface Type Message
hydra:hydra hydra:hydra hydra_peers peer
postgresql:database hydra:pg-database postgresql_client regular
postgresql:database-peers postgresql:database-peers postgresql_peers peer
postgresql:restart postgresql:restart rolling_op peer
postgresql:upgrade postgresql:upgrade upgrade peer
traefik-k8s:ingress hydra:public-ingress ingress regular
traefik-k8s:peers traefik-k8s:peers traefik_peers peer
Grafana, Loki, and Prometheus
The Hydra operator integrates with Canonical Observability Stack (COS) bundle. It comes with a Grafana dashboard as well as Loki and Prometheus alert rules for basic common scenarios. To integrate with the COS bundle, after you deploy it, you can run:
$ juju integrate hydra:grafana-dashboard grafana:grafana-dashboard
$ juju integrate hydra:metrics-endpoint prometheus:metrics-endpoint
$ juju integrate loki:logging hydra:log-proxy
Tear things down
To tear things down, remove the entire hydra model in juju with
juju destroy-model hydra