- MinIO exports Prometheus compatible data by default which is node centric as an authorized endpoint at `/minio/v2/metrics/node`.
- MinIO exports Prometheus compatible data by default which is resource centric as an authorized endpoint at `/minio/v2/metrics/resource`.

1. Start a distributed MinIO instance (4 nodes setup)
2. Start Prometheus server and AlertManager
3. Bring down couple of MinIO instances to bring down the Erasure Set tolerance to -1 and verify the same with `mc admin prometheus metrics ALIAS | grep minio_cluster_health_erasure_set_status`
4. Wait for 5 mins (as alert is configured to be firing after 5 mins), and verify that you see an entry in webhook for the alert as well as in Prometheus console as shown below

-   Continuous jobs (Run on every GitHub commit)
    -   Uses `pycpp.sh`
-   Presubmit jobs (Run on every GitHub PR)
    -   Uses `pycpp.sh`, `code_check_changed_files.sh`

These "env" files match up with an environment matrix that roughly covers:

-   Different Python versions
-   Linux, MacOS, and Windows machines (these pool definitions are internal)
-   x86 and arm64
-   CPU-only, or with NVIDIA CUDA support (Linux only), or with TPUs

The additional node specific metrics which include additional go metrics or process metrics are exposed at
`<Address for MinIO Node>/minio/v2/metrics/node`.

The additional bucket specific metrics which include additional go metrics or process metrics are exposed at
`<Address for MinIO Node>/minio/v2/metrics/bucket`.

The additional resource specific metrics which include additional go metrics or process metrics are exposed at

![Grafana](https://raw.githubusercontent.com/minio/minio/master/docs/metrics/prometheus/grafana/replication/grafana-replication-node.png)

- **Version Compatibility:**
With Java serialization, even minor changes to a class (like adding a field) can break compatibility.

- **Cross-Language Compatibility:**
Java serialization is inherently Java-centric and does not support cross-language scenarios well.

- **Type Safety:**
Java serialization does not enforce type safety as strictly as some alternatives, potentially leading to runtime errors.

## Decision