Skip to content

Commit

Permalink
Merge pull request #371 from device42/hotfix-zane-updates
Browse files Browse the repository at this point in the history
Hotfix zane updates
  • Loading branch information
sixhobbits authored Sep 21, 2024
2 parents cd62f25 + e4ce6d3 commit 7432032
Show file tree
Hide file tree
Showing 5 changed files with 25 additions and 24 deletions.
Original file line number Diff line number Diff line change
Expand Up @@ -33,6 +33,7 @@ Device42 discovery supports the following cloud types, as listed in the dropdown
/>

## Alibaba Cloud Autodiscovery
What does Alibaba Cloud Discovery gather? In addition to standard discovery methods, the discovery monitors network traffic to see all routes during discovery execution for Application Dependency Mapping. A read only Alibaba role can be provided as the discovery performs no changes to the cloud infrastructure during execution.

**Alibaba Discovery Items**

Expand Down
2 changes: 1 addition & 1 deletion docs/auto-discovery/enrichai-data/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -120,7 +120,7 @@ Please make sure the following is in place to ensure connectivity with EnrichAI:
- `https://enrichai.device42.io/api/normalizations`
3. Register the license online.

[DOQL AI](/reports/device42-doql/doql-ai.mdx) generates queries based on natural language prompts you provide. Entered prompts are sent to the endpoint `ai42.device42.io`.
[InsightsAI AI](/reports/device42-doql/doql-ai.mdx) generates queries based on natural language prompts you provide. Entered prompts are sent to the endpoint `ai42.device42.io`.

## Using the EnrichAI Interface

Expand Down
2 changes: 2 additions & 0 deletions docs/auto-discovery/load-balancers.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,6 +7,8 @@ sidebar_position: 18

Device42 SNMP autodiscovery provides a new option for F5 load balancer discovery that returns discovered load balancers as both devices and Managed Resources. Discovering F5 load balancers as managed resources provides additional details about the load balancers and related resources that include resource maps that visually show the relationships between load balancers and their related resources. It also simplifies the chart view for Affinity Groups by visually presenting the servers underneath a load balancer.

The recommended best practice for Load balancer discovery is to first identify the devices using SNMP. This will identify the devices and associated resources. Next perform a Windows/*nix discovery of the back end resources to get more detailed inforamtion. Once you have the devices discovered you can then perform the API based "Load Balancer" discovery to link the devices correctly.

## Managed Resources Views

Discovered F5 load balancers are included in the Managed Resources list. Select _Resources > Managed Resources_ from the Device42 menu to display the list. See the [Managed Resources](https://docs.device42.com/auto-discovery/resources/) page for more information.
Expand Down
2 changes: 1 addition & 1 deletion docs/auto-discovery/sccm-discovery.mdx
Original file line number Diff line number Diff line change
Expand Up @@ -12,7 +12,7 @@ As of 18.11.00, SCCM Discovery is integrated into our main discovery jobs. Micro

If you are already using SCCM in your environment, the Device42 SCCM integration can automatically sync the hardware and software inventory (CI) data to Device42. SCCM discovery can discover devices, OS, CPU, memory, network, and software from SCCM.

SCCM Discovery can be configured using either WinRM, WMI (Requires a WDS), or direct database discovery.
SCCM Discovery can be configured using either WinRM, WMI (Requires a WDS and read only access to the SMS namespace), or direct database discovery.

### View, Run, and Add Jobs

Expand Down
42 changes: 20 additions & 22 deletions docs/reports/device42-doql/doql-ai.mdx
Original file line number Diff line number Diff line change
@@ -1,31 +1,29 @@
---
title: "DOQL AI Chat"
title: "InsightsAI Chat"
sidebar_position: 3
---

import ThemedImage from '@theme/ThemedImage'
import useBaseUrl from '@docusaurus/useBaseUrl'

DOQL AI generates queries from the natural language descriptions you provide, streamlining the process of getting information and detailed reports from your Device42 CMDB.
InsightsAI allows users to generates queries from the natural language descriptions you provide, streamlining the process of getting information and detailed reports from your CMDB.

By typing plain English messages, like “I want a list of devices with Windows 10 operating system,” you'll get queries that can be executed without requiring deep knowledge of the query language or data model.

The AI provides the necessary commands and you can preview the results or create standard reports directly.
By typing plain English messages, like “Which SSL/TLS certificates are expiring in the next 90 days?” InsightsAI willl create the query that can be executed without requiring deep knowledge of the query language or data model. You are able to preview the results, and refine it as needed.

## Create a New Chat

Navigate to **Analytics > Ask DOQL AI** from the Device42 Main Appliance to create a new chat.
Navigate to **Analytics > Ask InsightsAI** from the Device42 Main Appliance to create a new chat.

<ThemedImage
alt="DOQL AI menu location"
alt="InsightsAI menu location"
sources={{
light: useBaseUrl('/assets/images/doql-ai/1-menu-light.png'),
dark: useBaseUrl('/assets/images/doql-ai/1-menu-dark.png'),
}}
style={{ width: '70%' }}
/>

From the Ask DOQL AI page, type in a description of the data you want to fetch in the text box.
From the Ask InsightsAI page, type in a description of the data you want to fetch in the text box.

<ThemedImage
alt="DOQL AI page"
Expand All @@ -49,11 +47,11 @@ Get me a query for 50 resources discovered in the past week ordered by resource

Click the **+ New Chat** button to create a new chat in a new browser tab.

Go to the DOQL AI Chats list page by clicking the **Past Chats** button.
Go to the InsightsAI Chats list page by clicking the **Past Chats** button.

## Query Result Options

DOQL AI generates an SQL query with the table and parameters in response to your input. See [Database Viewer Schema](db-viewer-schema.md) for information on locating database field names and descriptions.
InsightsAI generates an SQL query with the table and parameters in response to your input. See [Database Viewer Schema](db-viewer-schema.md) for information on locating database field names and descriptions.

The input for this example is:

Expand All @@ -71,9 +69,9 @@ find all devices in production

### Refine the Responses

If the data or query needs to be refined, ask a follow-up question or redirect the DOQL AI to get better or different data.
If the data or query needs to be refined, ask a follow-up question, InsightsAI will provide a new responsee.

For example, the previous results did not include information on device types. You can give DOQL AI another direction to include the type of device and only find virtual devices.
For example, the previous results did not include information on device types. You can give InsightsAI another direction to include the type of device and only find virtual devices.

<ThemedImage
alt="Follow-up response"
Expand All @@ -83,7 +81,7 @@ For example, the previous results did not include information on device types. Y
}}
/>

DOQL AI will run queries and self-correct if necessary. For example, when asked to look for Windows 10 and 11 devices, the AI encountered an error as it initially searched in the `os` column, but made the correction of looking in the `os_name` column.
InsightsAI will run queries and self-correct if necessary. For example, when asked to look for Windows 10 and 11 devices, the AI encountered an error as it initially searched in the `os` column, but made the correction of looking in the `os_name` column.

<ThemedImage
alt="Self-correction example"
Expand All @@ -93,11 +91,11 @@ DOQL AI will run queries and self-correct if necessary. For example, when asked
}}
/>

Although the DOQL AI brain knows the data model well, it continues to learn and train on the input and will provide increasingly accurate responses and need fewer refinement questions over time.
Although the InsightsAI brain knows the data model well, it continues to learn and train on the input and will provide increasingly accurate responses and need fewer refinement questions over time. InsightsAI only has access to the default database schema, not your data.

### Copy the SQL Response

From the DOQL AI response, select the **Copy** option to copy the SQL query to your clipboard. You can paste the SQL code into a new tab of an existing report to build a report.
From the InsightsAI response, select the **Copy** option to copy the SQL query to your clipboard. You can paste the SQL code into a new tab of an existing report to build a report.

### Preview Query Results

Expand All @@ -115,7 +113,7 @@ You can search for specific items using the **Search** bar, search individual co

### Create a Report Page

To save the generated query, click **Report** on the DOQL AI [response toolbar](#query-result-options) to go to the report setup page and provide a **Report Name**.
To save the generated query, click **Report** on the InsightsAI [response toolbar](#query-result-options) to go to the report setup page and provide a **Report Name**.

<ThemedImage
alt="Report page options"
Expand All @@ -138,14 +136,14 @@ To save the generated query, click **Report** on the DOQL AI [response toolbar](
- **(2) Export:** Download a `.tsv` or `.xls` file of the query result.
- **(3) Report Schedule:** Periodically run the query and email the results by filling in the fields.

## DOQL AI Chats List Page
## InsightsAI Chats List Page

Click the **Past Chats** button on the [Ask DOQL AI](#create-a-new-chat) page or [chat response](#query-result-options) page to navigate to the list page. The DOQL AI Chats List Page contains a record of your past chats that are automatically saved.
Click the **Past Chats** button on the [Ask DOQL AI](#create-a-new-chat) page or [chat response](#query-result-options) page to navigate to the list page. The InsightsAI Chats List Page contains a record of your past chats that are automatically saved.

You can create a **New Chat** from the list page.

<ThemedImage
alt="DOQL AI Chats list page"
alt="InsightsAI Chats list page"
sources={{
light: useBaseUrl('/assets/images/doql-ai/7-list-page-light.png'),
dark: useBaseUrl('/assets/images/doql-ai/7-list-page-dark.png'),
Expand Down Expand Up @@ -176,10 +174,10 @@ Delete one or more chats by selecting them and choosing a delete option from the

The following notice is available when you click **About DOQL AI** below the chat bar:

DOQL AI uses publicly available Device42 field names and descriptions for training and does not expose any data residing in the internal tables or data structures that include customer-specific or proprietary data. The field names and descriptions used for training the large language model and generative AI engine are documented and can be found at https://docs.device42.com/reports/device42-doql/db-viewer-schema/.
InsightsAI uses publicly available Device42 field names and descriptions for training and does not expose any data residing in the internal tables or data structures that include customer-specific or proprietary data. The field names and descriptions used for training the large language model and generative AI engine are documented and can be found at https://docs.device42.com/reports/device42-doql/db-viewer-schema/.

DOQL AI generates SQL query as a response to customer natural language inquiries. The inquiries are then uploaded to the DOQL AI cloud service, which generates the SQL response. DOQL AI utilizes OpenAI APIs as part of its generative AI process. The generated SQL is validated for accuracy and error free execution however the customer CMDB data is not exposed as part of the process.
InsightsAI generates SQL query as a response to customer natural language inquiries. The inquiries are then uploaded to the InsightsAI cloud service, which generates the SQL response. InsightsAI utilizes OpenAI APIs as part of its generative AI process. The generated SQL is validated for accuracy and error free execution however the customer CMDB data is not exposed as part of the process.

:::tip
See [Device42 Object Query Language](index.md) for information on DOQL queries.
See [Device42 Object Query Language](index.md) for information on database queries.
:::

0 comments on commit 7432032

Please sign in to comment.