---
type: slide
---

## Pulp Operator and High Availability

--

#### Dennis Kliban / dkliban

---

## Agenda

- What is High Availability (HA)?
- Architecture of Pulp
- Pulp Operator Responsibilities
- Demo of Pulp Operator scaling Pulp accross multiple Kubernetes nodes

---

## What is High Availability?

- Service Level Objectives (SLO)

- Service Level Indicators (SLI)

---

## Pulp Architecture

<div style="background-color: #FFFFFF; display: inline-block;">
<img src="https://docs.pulpproject.org/pulpcore/_images/architecture.png" width="75%">
</div>

---

## Pulp Operator Responsibilities

Ensure the right number of:
- pulpcore-api
- pulpcore-content
- pulpcore-workers

---

## What the operator DOES NOT do

- Ensure that the database is HA
- Ensure that the storage system is HA
- Ensure that the Redis cache is HA
- Ensure that the DNS services are HA
- Ensure that the reverse proxy is HA

---

# Demo

---

## Documentation
- https://docs.pulpproject.org/pulp_operator/install/ha/

- https://kubernetes.io/docs/concepts/scheduling-eviction/assign-pod-node/