Create a workload

A workload is your namespace on Plexivo Kubernetes, with the capacity its pods may use. Create one from the console, then keep working in the console or with kubectl.

Before you begin#

• An organisation on the Startup plan or above. See Plans and limits.
• A container image your pods can pull, with a tag or digest.

Deploy a container#

The quickest start. Plexivo writes the Kubernetes YAML for you.

1. Start. Choose Create, then Plexivo Kubernetes, and keep Deploy a container.
2. Container. Enter the Container image, the Workload name, the Project and the Container port your container listens on.
3. Size and scaling. Choose the Pod size and the number of Pods. Tick Scale automatically to set Minimum pods and Maximum pods; Plexivo adds pods when average CPU passes 70% of what each pod requests, and removes them when it falls.
4. Access. Tick Make it public on its own HTTPS address for an address Plexivo issues and renews the certificate for. Leave it off and the workload is private, reachable from your other Plexivo Kubernetes pods at http://<name>:80.
5. Environment variables. Add any you need. Keep passwords and keys for Plexivo Secrets, attached once the workload runs.
6. Review and deploy. Open See the YAML we generate to read exactly what is deployed, or choose Edit as YAML to add volumes, probes or more containers. Then check the price under What you pay for and choose the Deploy button, which carries the workload’s name.

A workload made this way has no persistent storage. For volumes, choose Edit as YAML and set Storage; see Storage.

Bring your own manifests#

Choose Advanced: YAML, Helm or Git, then one of:

• YAML: paste your manifests and review them.
• Helm chart: pick the Repository, Branch, Chart path and Release name, and optional Values.
• Git folder: pick the Repository, Branch and Manifests path; every .yaml under it is used.

For Helm and Git, choose Render for review. The rendered YAML lands in the editor so you see every object before it is deployed. Then set Capacity per pod (CPU, Memory, Storage) and Minimum pods and Maximum pods. Storage is what all the workload’s volume claims may request together.

Your manifests must follow a few rules, such as requests equal to limits. See Manifests and limits.

Pod sizes#

Plexivo fits the workload’s pod size to a few simple rules, and tells you before you deploy if it changed anything:

• At least 0.25 vCPU, in steps of 0.25 vCPU.
• At least 512 MB of memory.
• Between 1 GB and 6.5 GB of memory per vCPU.
• At most 16 vCPU and 64 GB per pod.

After it is created#

The workload’s Overview shows CPU, memory and storage in use, its controllers and pods, its public HTTPS addresses and recent warnings. Connect has your kubeconfig and database connections. Metrics shows live usage and its history. Logs & events has pod logs and Kubernetes events. Settings changes the requested capacity, attaches secrets and deletes the workload.

If the region has no room for the workload at that moment, nothing is created or charged, and the console asks you to try again in a few minutes.

Next steps#