Project Quotas
Each project on Bahriya can have resource quotas that limit the total CPU and memory available to all containers within that project. Quotas help prevent a single project from consuming more resources than intended.
What quotas control
| Quota | Description |
|---|
| Request CPU | Total CPU that can be requested across all containers in the project |
| Request memory | Total memory that can be requested across all containers in the project |
| Limit CPU | Total CPU limit across all containers in the project |
| Limit memory | Total memory limit across all containers in the project |
CPU is measured in millicores (e.g. 1000m = 1 full CPU core). Memory is measured in mebibytes (e.g. 2048Mi = 2 GB).
How quotas interact with containers
When you create or update a container, you set its minimum CPU and memory (the baseline it needs) and Bahriya calculates the limit (the maximum it can burst to). The sum of all container resource settings in a project must fit within the project's quotas.
If you try to create a container or increase resources beyond the quota, the operation will be rejected.
How quotas are set
Quotas are set by the Bahriya team or your organisation's account manager. They are not self-service. If you need higher quotas, contact support with details about your expected usage.
Checking your quotas
You can see your project's current quotas and usage in the project settings page in the console.
What happens when you hit a quota
- Creating a new container: The creation is rejected with an error indicating which quota would be exceeded.
- Scaling up: If autoscaling would push total resource usage beyond the quota, new replicas are not created.
- Updating resources: Increasing CPU or memory on an existing container is rejected if it would exceed the quota.
Your existing containers continue to run normally — quotas only prevent new allocations that would exceed the limit.
Tips
- Check your quota usage before creating new containers, especially in projects with many services.
- If you are running close to your quota and expect to need more capacity, request a quota increase before you need it — it avoids deployment failures during traffic spikes.
