Since SQLite is a single-file database, usually only the host where the database is located can access it. Therefore, SQLite database is more suitable for stand-alone use. For multiple servers sharing the same file system, it is recommended to use databases such as Redis or MySQL.
Here we assign AWS [IAM role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_use_switch-role-ec2.html) for the EC2 Kuberentes node, otherwise the `accessKey` and `secretKey` cannot be empty. We use ElastiCache for Redis as the meta store.
Here we assign AWS [IAM role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_use_switch-role-ec2.html) for the EC2 Kubernetes node, otherwise the `accessKey` and `secretKey` cannot be empty. We use ElastiCache for Redis as the meta store.
Here we use the `juicedata/juicefs-csi-driver:latest` image, if we want to use the specified tag such as `v0.7.0` , we should download the deploy YAML file and modified it:
Here we use the `juicedata/juicefs-csi-driver:latest` image, if you want to use the specified tag such as `v0.7.0`, you should download the deploy YAML file and modified it:
If the status of pod is not `Running` (e.g. `ContainerCreating`), there may have some issues. Please refer to the [troubleshooting](https://github.com/juicedata/juicefs-csi-driver/blob/master/docs/troubleshooting.md) document.
For more details about JuiceFS CSI driver please refer to [JuiceFS CSI Driver](https://github.com/juicedata/juicefs-csi-driver).
For more details about JuiceFS CSI Driver please refer to [project homepage](https://github.com/juicedata/juicefs-csi-driver).
## Monitoring
JuiceFS CSI driver can export [Prometheus](https://prometheus.io) metrics at port `9567`. For a description of all monitoring metrics, please refer to [JuiceFS Metrics](p8s_metrics.md).
JuiceFS CSI Driver can export [Prometheus](https://prometheus.io) metrics at port `9567`. For a description of all monitoring metrics, please refer to [JuiceFS Metrics](p8s_metrics.md).
### Configure Prometheus server
...
...
@@ -209,7 +211,7 @@ scrape_configs:
action:replace
```
Here we assume the Prometheus server is running inside Kubernetes cluster, if your prometheus server is running outside Kubernetes cluster, make sure Kubernetes cluster nodes are reachable from Prometheus server, refer to [this issue](https://github.com/prometheus/prometheus/issues/4633) to add the `api_server` and `tls_config` client auth to the above configuration like this:
Here we assume the Prometheus server is running inside Kubernetes cluster, if your Prometheus server is running outside Kubernetes cluster, make sure Kubernetes cluster nodes are reachable from Prometheus server, refer to [this issue](https://github.com/prometheus/prometheus/issues/4633) to add the `api_server` and `tls_config` client auth to the above configuration like this:
```yaml
scrape_configs:
...
...
@@ -229,4 +231,4 @@ scrape_configs:
### Configure Grafana dashboard
We provide a [dashboard template](./k8s_grafana_template.json) for [Grafana](https://grafana.com), which can be imported to show the collected metrics in Prometheus.
JuiceFS provides a [dashboard template](./k8s_grafana_template.json) for [Grafana](https://grafana.com), which can be imported to show the collected metrics in Prometheus.
Here we assign AWS [IAM role](https://docs.aws.amazon.com/IAM/latest/UserGuide/id_roles_use_switch-role-ec2.html) for the EC2 Kuberentes node, otherwise the `accessKey` and `secretKey` cannot be empty. We use ElastiCache for Redis as the meta store.
-Check pods are running: the deployment will launch a `StatefulSet` named `juicefs-csi-controller` with replica `1` and a `DaemonSet` named `juicefs-csi-node`, so run `kubectl -n kube-system get pods -l app.kubernetes.io/name=juicefs-csi-driver` should see `n+1` (where `n` is the number of worker nodes of the Kubernetes cluster) pods is running. For example:
Here we use the `juicedata/juicefs-csi-driver:latest` image, if we want to use the specified tag such as `v0.7.0` , we should download the deploy YAML file and modified it:
Now we can use JuiceFS in our pods. Here we create a `PVC` and refer to it in a pod as an example:
现在我们可以在 pods 中使用 JuiceFS。这里我们创建了一个 `PersistentVolumeClaim` 并在一个 pod 中使用它：
```yaml
apiVersion:v1
...
...
@@ -169,7 +171,7 @@ spec:
claimName:juicefs-pvc
```
Save above content to a file named like `juicefs-app.yaml`, then use command `kubectl apply -f juicefs-app.yaml` to bootstrap the pod. After that, you can check the status of pod:
@@ -177,18 +179,18 @@ NAME READY STATUS RESTARTS AGE
juicefs-app 1/1 Running 0 10m
```
If the status of pod is not `Running` (e.g. `ContainerCreating`), there may have some issues. Please refer to the [troubleshooting](https://github.com/juicedata/juicefs-csi-driver/blob/master/docs/troubleshooting.md) document.
如果 pod 的状态不是 `Running`（例如 `ContainerCreating`），应该是存在一些问题。请参考[故障处理](https://github.com/juicedata/juicefs-csi-driver/blob/master/docs/troubleshooting.md)文档。
For more details about JuiceFS CSI driver please refer to [JuiceFS CSI Driver](https://github.com/juicedata/juicefs-csi-driver).
JuiceFS CSI driver can export [Prometheus](https://prometheus.io) metrics at port `9567`. For a description of all monitoring metrics, please refer to [JuiceFS Metrics](p8s_metrics.md).
Here we assume the Prometheus server is running inside Kubernetes cluster, if your prometheus server is running outside Kubernetes cluster, make sure Kubernetes cluster nodes are reachable from Prometheus server, refer to [this issue](https://github.com/prometheus/prometheus/issues/4633) to add the `api_server` and `tls_config` client auth to the above configuration like this:
We provide a [dashboard template](../en/k8s_grafana_template.json) for [Grafana](https://grafana.com), which can be imported to show the collected metrics in Prometheus.
