Enable and use the Fluid JindoRuntime FUSE client for monitoring

Fluid provides an out-of-the-box JindoRuntime dashboard that collects monitoring metrics for each JindoCache distributed cache engine in your ACK cluster. The dashboard covers two metric categories: server-side cache engine metrics and Filesystem in Userspace (FUSE) client metrics. For performance reasons, FUSE client metrics are not exposed by default, so they do not appear on the dashboard until you explicitly enable them.

This topic describes how to enable FUSE client metrics collection and view the data on the Fluid JindoRuntime dashboard. After you complete the three steps below, FUSE client metrics appear in dedicated sections on the dashboard — FUSE Metrics (via CSI) for MountPod mode and FUSE Metrics (via Sidecar) for Sidecar mode.

Prerequisites

Before you begin, ensure that you have:

Managed Service for Prometheus enabled in your ACK cluster or ACK Serverless cluster. For more information, see Use Managed Service for Prometheus
The cloud-native AI suite and ack-fluid version 1.0.11 or later installed in your cluster. For more information, see Deploy the cloud-native AI suite

Important

If you have the open-source Fluid installed, uninstall it before installing ack-fluid to prevent conflicts.

Step 1: Integrate Fluid with ARMS

Log on to the ARMS console.
In the left-side navigation pane, click Integration Center. In the AI section, click the Fluid card.
In the Select a Kubernetes cluster section, select your cluster. If the page shows that Fluid is already installed, skip the remaining steps in this section.
In the Configuration Information section, set the following parameters and click OK.
Parameter Description
Name (Not required) A unique name for the Fluid exporter. Leave blank if not needed.
metrics collection interval (seconds) The interval at which the service collects monitoring data.
To verify the integration, go to the Integration Management page of the ARMS console.
1. Log on to the ARMS console.
2. In the left-side navigation pane, click Integration Management. On the Integrated Addons tab, click the Fluid card.
3. On the Environments tab, click View Details in the Actions column to view the Fluid component status and alert rules in your cluster.

Parameter	Description
Name (Not required)	A unique name for the Fluid exporter. Leave blank if not needed.
metrics collection interval (seconds)	The interval at which the service collects monitoring data.

Step 2: Configure metrics exposure and scraping policy

When deploying JindoRuntime, use the spec.fuse.metrics.enabled and spec.fuse.metrics.scrapeTarget fields to enable FUSE client metrics exposure and set the scraping policy.

Choose your scraping policy before writing the YAML. The scrapeTarget value determines which dashboard sections are populated:

Use MountPod if your workloads run in non-Serverless environments using MountPod mount mode.
Use Sidecar if your workloads run in Serverless environments using Sidecar mount mode.
Use All to scrape from both mount modes.
Use None (default) to disable FUSE client metrics scraping.

Warning

Set scrapeTarget to Sidecar or All with caution in Serverless environments. In Sidecar mount mode, each pod starts a FUSE sidecar container. When the number of concurrent pods is high, scraping metrics from all FUSE sidecar containers can create significant overhead and place considerable pressure on the Managed Service for Prometheus instance.

The following YAML shows an example with scrapeTarget: All:

apiVersion: data.fluid.io/v1alpha1
kind: JindoRuntime
metadata:
  name: hadoop
spec:
  replicas: 2
  fuse:
    metrics:
      enabled: true
      scrapeTarget: All
  tieredstore:
    levels:
      - mediumtype: MEM
        path: /dev/shm
        volumeType: emptyDir
        quota: 2Gi
        high: "0.99"
        low: "0.95"

The following table describes the two metrics-related parameters:

Parameter	Description	Default value
`spec.fuse.metrics.enabled`	Whether to expose FUSE client metrics. When set to `true` without a port specified via `spec.fuse.args` (for example, `-ometrics_port=8080`), Fluid automatically assigns an available port to the FUSE client and configures it for external metrics services.	`false`
`spec.fuse.metrics.scrapeTarget`	The scraping policy for FUSE client metrics. Supported values: `None`, `MountPod`, `Sidecar`, `All`. See the decision guidance above for when to use each value.	`None`

Note

For the full list of JindoRuntime parameters, see Use JindoFS to accelerate access to OSS.

Step 3: View the Fluid JindoRuntime dashboard

Log on to the ACK console. In the left-side navigation pane, click Clusters.
On the Clusters page, click the name of your cluster. In the left-side pane, choose Operations > Prometheus Monitoring.
On the Prometheus Monitoring page, click the Others tab, then click the Fluid JindoRuntime Dashboard tab. The dashboard displays FUSE client metrics in the following sections, depending on your scrapeTarget configuration:
- FUSE Metrics (via CSI): Displays metrics from FUSE pod clients. Populated when scrapeTarget is set to MountPod or All.
- FUSE Metrics (via Sidecar): Displays metrics from FUSE sidecar container clients. Populated when scrapeTarget is set to Sidecar or All.

Container Service for Kubernetes:Enable and use the Fluid JindoRuntime FUSE client for monitoring

Prerequisites

Step 1: Integrate Fluid with ARMS

Step 2: Configure metrics exposure and scraping policy

Step 3: View the Fluid JindoRuntime dashboard

What's next