From 3a668b02f4fcae4035cf8186ce138acc9aef925a Mon Sep 17 00:00:00 2001 From: ruquanzhao <903264308@qq.com> Date: Wed, 15 Jan 2025 12:36:50 +0800 Subject: [PATCH] docs: add enable AI feature docs of Karpor --- .../1-getting-started/2-installation.md | 41 +++++++++++++++++-- .../karpor/1-getting-started/3-quick-start.md | 5 ++- .../1-getting-started/2-installation.md | 27 ++++++++++++ .../1-getting-started/3-quick-start.md | 5 ++- 4 files changed, 71 insertions(+), 7 deletions(-) diff --git a/docs/karpor/1-getting-started/2-installation.md b/docs/karpor/1-getting-started/2-installation.md index 661d99a1..6bdc2696 100644 --- a/docs/karpor/1-getting-started/2-installation.md +++ b/docs/karpor/1-getting-started/2-installation.md @@ -89,6 +89,33 @@ helm install karpor-release kusionstack/karpor --set registryProxy=docker.m.daoc **NOTE**: The above is just an example, you can replace the value of `registryProxy` as needed. +### Enable AI features + +If you are trying to install Karpor with AI features, including natural language search and AI analyze, `ai-auth-token` and `ai-base-url` should be configured, e.g.: + +```shell +# At a minimum, server.ai.authToken and server.ai.baseUrl must be configured. +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 +# server.ai.backend has default values `openai`, which can be overridden when necessary. If the backend you are using is compatible with OpenAI, then there is no need to make any changes here. +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.backend=huggingface +# server.ai.model has default values `gpt-3.5-turbo`, which can be overridden when necessary. +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.model=gpt-4o +# server.ai.topP and server.ai.temperature can also be manually modified. +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.topP=0.5 \ +--set server.ai.temperature=0.2 +``` + ### Chart Parameters The following table lists the configurable parameters of the chart and their default values. @@ -113,13 +140,21 @@ The Karpor Server Component is main backend server. It itself is an `apiserver`, | Key | Type | Default | Description | |-----|------|---------|-------------| +| server.ai | object | `{"authToken":"","backend":"openai","baseUrl":"","model":"gpt-3.5-turbo","temperature":1,"topP":1}` | AI configuration section. The AI analysis feature requires that [authToken, baseUrl] be assigned values. | +| server.ai.authToken | string | `""` | Authentication token for accessing the AI service. | +| server.ai.backend | string | `"openai"` | Backend service or platform that the AI model is hosted on. e.g., "openai". If the backend you are using is compatible with OpenAI, then there is no need to make any changes here. | +| server.ai.baseUrl | string | `""` | Base URL of the AI service. e.g., "https://api.openai.com/v1". | +| server.ai.model | string | `"gpt-3.5-turbo"` | Name or identifier of the AI model to be used. e.g., "gpt-3.5-turbo". | +| server.ai.temperature | float | `1` | Temperature parameter for the AI model. This controls the randomness of the output, where a higher value (e.g., 1.0) makes the output more random, and a lower value (e.g., 0.0) makes it more deterministic. | +| server.ai.topP | float | `1` | Top-p (nucleus sampling) parameter for the AI model. This controls Controls the probability mass to consider for sampling, where a higher value leads to greater diversity in the generated content (typically ranging from 0 to 1) | +| server.enableRbac | bool | `false` | Enable RBAC authorization if set to true. | | server.image.repo | string | `"kusionstack/karpor"` | Repository for Karpor server image. | | server.image.tag | string | `""` | Tag for Karpor server image. Defaults to the chart's appVersion if not specified. | | server.name | string | `"karpor-server"` | Component name for karpor server. | | server.port | int | `7443` | Port for karpor server. | | server.replicas | int | `1` | The number of karpor server pods to run. | | server.resources | object | `{"limits":{"cpu":"500m","ephemeral-storage":"10Gi","memory":"1Gi"},"requests":{"cpu":"250m","ephemeral-storage":"2Gi","memory":"256Mi"}}` | Resource limits and requests for the karpor server pods. | -| server.serviceType | string | `"ClusterIP"` | Service type for the karpor server. The available type values list as ["ClusterIP"、"NodePort"、"LoadBalancer"]. | +| server.serviceType | string | `"ClusterIP"` | Service type for the karpor server. The available type values list as ["ClusterIP"、"NodePort"、"LoadBalancer"]. | #### Karpor Syncer @@ -156,8 +191,8 @@ The ETCD Component is the storage of Karpor Server as `apiserver`. | etcd.image.repo | string | `"quay.io/coreos/etcd"` | Repository for ETCD image. | | etcd.image.tag | string | `"v3.5.11"` | Specific tag for ETCD image. | | etcd.name | string | `"etcd"` | Component name for ETCD. | -| etcd.persistence.accessModes[0] | string | `"ReadWriteOnce"` | | -| etcd.persistence.size | string | `"10Gi"` | | +| etcd.persistence.accessModes | list | `["ReadWriteOnce"]` | Volume access mode, ReadWriteOnce means single node read-write access | +| etcd.persistence.size | string | `"10Gi"` | Size of etcd persistent volume | | etcd.port | int | `2379` | Port for ETCD. | | etcd.replicas | int | `1` | The number of etcd pods to run. | | etcd.resources | object | `{"limits":{"cpu":"500m","ephemeral-storage":"10Gi","memory":"1Gi"},"requests":{"cpu":"250m","ephemeral-storage":"2Gi","memory":"256Mi"}}` | Resource limits and requests for the karpor etcd pods. | diff --git a/docs/karpor/1-getting-started/3-quick-start.md b/docs/karpor/1-getting-started/3-quick-start.md index 4d3839c6..23afc98b 100644 --- a/docs/karpor/1-getting-started/3-quick-start.md +++ b/docs/karpor/1-getting-started/3-quick-start.md @@ -119,5 +119,6 @@ If you have any questions or concerns, check out the official documentation of K ## Next Step -- Learn Karpor's [Architecture](../concepts/architecture) and [Glossary](../concepts/glossary). -- View [User Guide](../user-guide/multi-cluster-management) to look on more of what you can achieve with Karpor. +* Learn Karpor's [Architecture](../concepts/architecture) and [Glossary](../concepts/glossary). +* View [User Guide](../user-guide/multi-cluster-management) to look on more of what you can achieve with Karpor. +* [Enable AI features](installation#enable-ai-features), including natural language search and AI analyze. diff --git a/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/2-installation.md b/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/2-installation.md index 596a90ec..deb9fdc4 100644 --- a/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/2-installation.md +++ b/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/2-installation.md @@ -89,6 +89,33 @@ helm install karpor-release kusionstack/karpor --set registryProxy=docker.m.daoc **注意**: 以上只是一个样例，你可以根据需要替换 `registryProxy` 的值。 +### 启用 AI 功能 + +如果您要安装带有AI功能的Karpor，包括自然语言搜索和AI分析，则应配置 `ai-auth-token` 和 `ai-base-url`，例如： + +```shell +# 至少需要配置 server.ai.authToken 和 server.ai.baseUrl。 +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 +# server.ai.backend 的默认值是 `openai`，可以根据需要进行覆盖。如果你使用的后端与 OpenAI 兼容，则无需在此处进行任何更改。 +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.backend=huggingface +# server.ai.model 的默认值是 `gpt-3.5-turbo`，可以根据需要进行覆盖。 +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.model=gpt-4o +# server.ai.topP 和 server.ai.temperature 也可以手动修改。 +helm install karpor-release kusionstack/karpor \ +--set server.ai.authToken=YOUR_AI_TOKEN \ +--set server.ai.baseUrl=https://api.openai.com/v1 \ +--set server.ai.topP=0.5 \ +--set server.ai.temperature=0.2 +``` + ### Chart 参数 以下表格列出了 Chart 的所有可配置参数及其默认值。 diff --git a/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/3-quick-start.md b/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/3-quick-start.md index a669e4b7..ff8f35a2 100644 --- a/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/3-quick-start.md +++ b/i18n/zh/docusaurus-plugin-content-docs-karpor/current/1-getting-started/3-quick-start.md @@ -117,5 +117,6 @@ Karpor 提供了一个强大的搜索功能，允许你快速跨集群查找资 ## 下一步 -- 了解 Karpor 的 [架构](../concepts/architecture) 和 [术语表](../concepts/glossary)。 -- 查看 [用户指南](../user-guide/multi-cluster-management) 以了解 Karpor 的更多功能。 +* 了解 Karpor 的 [架构](../concepts/architecture) 和 [术语表](../concepts/glossary)。 +* 查看 [用户指南](../user-guide/multi-cluster-management) 以了解 Karpor 的更多功能。 +* [启用 AI 功能](installation#启用-ai-功能)，包括自然语言搜索和 AI 智能分析。