Opni = AIOps for Kubernetes + Observability Tools

Opni currently features log anomaly detection for Kubernetes.

What does Opni give me?

• AI generated insights on your cluster's log messages
  • Control Plane & etcd insights
    • Pretrained models maintained by Rancher Labs
• Every log message sent to Opni will be marked as:
  • Normal
  • Suspicious - Operators may want to investigate
  • Anomalous - Operators definitely should investigate
• Opensearch + Opensearch Dashboards
  • Opni dashboard to consume log insights & explore logs

Deprecation Notice

• GPU Learning is temporarily disabled in the v0.4.0 release as Opni moves to a multicluster architecture. This will be returning in a future release
• The v1beta1 API has been deprecated in this release. Please migrate to v1beta2.
• The UI and Insights services, which were experimental, have been removed

Getting started with Opni

Full Install Opni in your Kubernetes cluster:

Manifests install (recommended):

Prerequisites:

• Cert manager installed. This can be installed with the following command:

  kubectl apply -f https://github.com/cert-manager/cert-manager/releases/download/v1.7.2/cert-manager.yaml
  

• Opni Gateway installed - see the Main Cluster docs for Opni Monitoring

Installation:

1. All clusters (both the main cluster and clusters to collect logs from) the manifests in deploy/manifests in order from 00 - 10.
2. Deploy an Opensearch cluster e.g (node this cluster will need to be exposed via a LoadBalancer or Ingress to allow logs to be indexed)

   apiVersion: opensearch.opster.io/v1
   kind: OpenSearchCluster
   metadata:
     name: opni
     namespace: opni-cluster-system
   spec:
     # Add fields here
     general:
       httpPort: 9200
       vendor: opensearch
       version: 1.2.3
       serviceName: os-svc
       setVMMaxMapCount: true
     confMgmt:
       autoScaler: false
       monitoring: false
     dashboards:
       enable: true
       version: 1.2.0
       replicas: 1
     nodePools:
     - component: master
       replicas: 3
       diskSize: 32
       resources:
         requests:
           cpu: 500m
           memory: 1Gi
         limits:
           memory: 1Gi
       roles:
       - master
       persistence:
         emptyDir: {}
     - component: nodes
       replicas: 2
       diskSize: 32
       resources:
         requests:
           cpu: 500m
           memory: 2Gi
         limits:
           memory: 2Gi
       jvm: "-Xmx1G -Xms1G"
       roles:
       - data
       persistence:
         emptyDir: {}
   

3. Bind Opni to the Opensearch cluster:

   apiVersion: opni.io/v1beta2
   kind: MulticlusterRoleBinding
   metadata:
     name: opni-logging
     namespace: opni-cluster-system
   spec:
     opensearch:
       name: opni
       namespace: opni-cluster-system
     opensearchExternalURL: https://external.opensearch.url
   

4. Deploy the Opni pretrained Kubernetes model

   apiVersion: opni.io/v1beta2
   kind: PretrainedModel
   metadata:
     name: control-plane
     namespace: opni-cluster-system
   spec:
     source:
       http:
         url: "https://opni-public.s3.us-east-2.amazonaws.com/pretrain-models/control-plane-model-v0.4.0.zip"
     hyperparameters:
       modelThreshold: "0.6"
       minLogTokens: 1
       isControlPlane: "true"
   

5. Deploy Opni AI services

   apiVersion: opni.io/v1beta2
   kind: OpniCluster
   metadata:
     name: demo
     namespace: opni-cluster-system
   spec:
     version: v0.4.0
     deployLogCollector: false
     services:
       gpuController:
         enabled: false
       inference:
         pretrainedModels:
         - name: control-plane
     opensearch:
       externalOpensearch:
         name: opni
         namespace: opni-cluster-system
       enableLogIndexManagement: false
     s3:
       internal: {}
     nats:
       authMethod: nkey
   

6. Add additional Logging clusters from the Opni Gateway UI

Consume insights from the Opni Dashboard in Opensearch Dashboards. You will need to expose the Dashboards service or port forward to do this.

Watch a demo of Opni:

What's next?

• v0.1.1 (Released) allows you to view Opni's log anomaly insights only on a demo environment created on a VM
• v0.1.2 (Released) allows you install Opni into your existing Kubernetes cluster and consume log insights from it
• v0.1.3 (August 2021) - only 1 GPU required, changes to the Opni operator, log anomaly optimizations
• v0.2.0 (Fall 2021) will introduce a custom UI, AI applied to metrics, kubernetes events, audit logs, and more!

License

Copyright (c) 2014-2020 Rancher Labs, Inc.

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.

Opni Monitoring is an open-source multi-cluster monitoring system. It ingests Prometheus metrics from any number of Kubernetes clusters and provides a centralized observability plane for your infrastructure. Use Opni Monitoring to visualize metrics from all your clusters at once, and give every user their own customized view using granular access control.

⚡ Powered by Open-Source

Opni Monitoring is completely free Apache-licensed open-source software. It builds upon existing, ubiquitous open-source systems - Prometheus, Grafana, and Cortex - and extends them with a number of powerful enterprise features typically only found in SaaS platforms and other proprietery solutions.

🔋 Batteries Included

Opni Monitoring comes out of the box with all the tools you need to get started with multi-cluster monitoring. Manage your clusters and configure access control rules with the built-in dashboard, command-line interface, or REST API.

Opni Monitoring is secure-by-default and uses a zero-trust architecture for inter-cluster communication, with no extra setup required.

🔒 You Own Your Data

With Opni Monitoring, you have complete control over how and where your data is stored. Metric storage is powered by Cortex, which provides comprehensive configuration options for data storage and retention. Several storage backends are available including S3 (cloud or self-hosted), Swift, and Kubernetes Persistent Volumes.

Get started

Check out the Opni Monitoring Documentation for installation guides and more.