Tutorial: How to Create a Development Environment with Realistic Data in Okteto #959
Conversation
huguestennier
requested review from
codyjlandstrom,
pchico83,
ifbyol and
rlamana
as code owners
❌ Deploy Preview for okteto-docs failed.
|
There was a problem hiding this comment.
Looks great! My main suggestion is to use okteto deploy instead of kubectl commands to deploy the environment. okteto/fastapi-snapshot-app#2 has those changes, I just tested it and it works
| Using Okteto Volume Snapshots, you can quickly create development environments that include production or staging data—without manually copying databases or waiting for CI/CD pipelines. | ||
|
|
||
| :::tip | ||
| Volume Snapshots are also useful for loading test data when running tests in Okteto Test. If your tests require a database with preloaded data, you can configure your test environment to restore from a snapshot instead of initializing an empty database. This ensures consistency across test runs and reduces setup time. |
There was a problem hiding this comment.
we should do a benchmark at some point but I wouldn't be surprise is there's a 5-10X difference
| Volume Snapshots are also useful for loading test data when running tests in Okteto Test. If your tests require a database with preloaded data, you can configure your test environment to restore from a snapshot instead of initializing an empty database. This ensures consistency across test runs and reduces setup time. | |
| Volume Snapshots are also useful for loading test data when running tests in Okteto Test. If your tests require a database with preloaded data, you can configure your test environment to restore from a snapshot instead of initializing an empty database. This ensures consistency across test runs and significantly reduces setup time. |
| ```console | ||
| okteto kubeconfig | ||
| kubectl apply -f manifests/ | ||
| ``` |
There was a problem hiding this comment.
| ```console | |
| okteto kubeconfig | |
| kubectl apply -f manifests/ | |
| ``` | |
| ```console | |
| okteto deploy |
|
|
||
| Expected output: | ||
| ```console | ||
| deployment.apps/fastapi-snapshot-demo created |
There was a problem hiding this comment.
How about we put the okteto output? (I can send a PR to the sample to use okteto.yaml and deploy)
| deployment.apps/fastapi-snapshot-demo created | |
| i Using clone-wars @ demo.okteto.dev as context | |
| i Running stage 'kubectl apply -f manifests' | |
| Executing command 'kubectl apply -f manifests'... | |
| deployment.apps/fastapi-snapshot-demo unchanged | |
| service/fastapi-snapshot-demo unchanged | |
| persistentvolumeclaim/data unchanged | |
| service/mongodb unchanged | |
| statefulset.apps/mongodb unchanged | |
| Command 'kubectl apply -f manifests' successfully executed | |
| i Endpoints available: | |
| - https://fastapi-snapshot-demo-cindy.example.okteto.com | |
| ✓ Development environment 'fastapi-snapshot-app' successfully deployed |
| "ingredients": ["Flour", "Milk", "Sugar", "Vegetable Oil"] | ||
| }' | ||
| ``` | ||
|
|
There was a problem hiding this comment.
nit: how about a first step to put the endpoint on an env var, and then we use that for all the curls?
|
|
||
| A Volume Snapshot will capture the database state at this moment. | ||
|
|
||
| Create a new manifest file `snapshot.yml` in the `manifests` folder and add the following to it: |
There was a problem hiding this comment.
this is currently available at the root of the repo, its the vol.yaml file (I think). should we remove it? (or maybe we keep a separate branch with all the completed manifests?
| spec: | ||
| volumeSnapshotClassName: okteto-snapshot-class | ||
| source: | ||
| persistentVolumeClaimName: data-mongodb-0 |
There was a problem hiding this comment.
today, the sample creates a PVC called data. do you want me to update it so it's called data-mongodb-0 instead?
| - A **Kubernetes cluster with Okteto** installed | ||
| - Familiarity with **Kubernetes manifests** | ||
| - [kubectl](https://kubernetes.io/docs/tasks/tools/) installed on your machine |
There was a problem hiding this comment.
you also need to create the volumesnapshotclass that we use (https://kubernetes.io/docs/concepts/storage/volume-snapshot-classes/). I believe we have docs for this, I don't think we create one by default when you install Okteto (which maybe we should?)
| ### 1. Create a New Namespace for Development | ||
|
|
||
| ```console | ||
| okteto namespace create data-clone-$YOUR_OKTETO_USERNAME |
There was a problem hiding this comment.
since we are no longer multitenant, the suffix is not needed by default. maybe we don't include it to keep it simpler?
| cp manifests/{k8s.yml,mongo.yml} data-clone-manifest | ||
| ``` | ||
|
|
||
| Update `mongo.yml` to reference the snapshot: |
There was a problem hiding this comment.
the mongo.yml on the repo is diferent, it declares the volumeclaim first and then it uses it on the stateful set. do you want to change the sample to this model?
|
|
||
| ```console | ||
| mkdir data-clone-manifest | ||
| cp manifests/{k8s.yml,mongo.yml} data-clone-manifest |
There was a problem hiding this comment.
what if we include the manifests already on a separate folder ? or do you think there's value on the user manually modifying it as a learning experience?
Port of https://deploy-preview-1230--okteto-website.netlify.app/blog/how-to-create-and-use-data-clones-in-okteto-cloud/