JSM Agent and JSM Users are unable to search the KB articles from the HelpCenter of Customer Portal.
Platform Notice: Data Center - This article applies to Atlassian products on the Data Center platform.
Note that this knowledge base article was created for the Data Center version of the product. Data Center knowledge base articles for non-Data Center-specific features may also work for Server versions of the product, however they have not been tested. Support for Server* products ended on February 15th 2024. If you are running a Server product, you can visit the Atlassian Server end of support announcement to review your migration options.
*Except Fisheye and Crucible
Summary
Search for KB articles from HelpCenter doesn't returns any suggestions and lists only Request Types.
Environment
JSM DC 5.12.15
Confluence Cloud
Diagnosis
When a KB article is searched from the HelpCenter page, 2 requests are triggered that can be observed from the HAR file.
- It looks for the KB articles using the keywords provided by the Users.
- It looks for the Request Types and list the top 5 results as explained here.
Search 1
Request URL:https://<BaseURL>/rest/servicedesk/knowledgebase/latest/articles/shared/search?resultsPerPage=3&pageNumber=1&query=%255BTroubleshoot%255D%2520Forgot%2520password%2520for%2520Agile
Request Method:GET
Status Code:200{
"results": [],
"total": 0,
"pageNumber": 0,
"resultsPerPage": 0,
"isCloud": false
}Search 2
Request URL:https://<BaseURL>/rest/servicedesk/search/1/search/request-type?query=%255BTroubleshoot%255D%2520Forgot%2520password%2520for%2520Agile&resultLimit=5
Request Method:GET
Status Code:200{
"requestTypes": [
{
"id": 196,
"cvId": 4,
"icon": 10638,
"issueType": 10016,
"name": "Password Reset",
"portalName": "IT- I Need Something",
"description": "Request related to - Password Reset",
"descriptionContent": "<p>Request related to - Password Reset</p>",
"searchScore": 0,
"groups": [
{
"id": 33,
"name": "Password Reset",
"order": 18
}
]
},
{
"id": 94,
"cvId": 1,
"icon": 10630,
"issueType": 10000,
"name": "Something Is Broken",
"portalName": "IT- Something Is Broken",
"description": "I need help with an IT issue",
"descriptionContent": "<p>I need help with an IT issue</p>",
"searchScore": 0,
"groups": [
{
"id": 7,
"name": "General IT - Something Is Broken",
"order": 0
}
]
},
{
"id": 111,
"cvId": 4,
"icon": 10629,
"issueType": 10009,
"name": "Make a Request",
"portalName": "IT- I Need Something",
"description": "Request related to Make a Request",
"descriptionContent": "<p>Request related to Make a Request</p>",
"searchScore": 0,
"groups": [
{
"id": 23,
"name": "Application - Others",
"order": 2
}
]
},
{
"id": 141,
"cvId": 4,
"icon": 10629,
"issueType": 10009,
"name": "Agile",
"portalName": "IT- I Need Something",
"description": "Request related to Agile",
"descriptionContent": "<p>Request related to Agile</p>",
"searchScore": 0,
"groups": [
{
"id": 23,
"name": "Application - Others",
"order": 2
}
]
},
{
"id": 751,
"cvId": 4,
"icon": 10624,
"issueType": 10015,
"name": "Inquiry",
"portalName": "IT- I Need Something",
"description": "Request related to Service Desk Queries",
"descriptionContent": "<p>Request related to Service Desk Queries</p>",
"searchScore": 0,
"groups": [
{
"id": 116,
"name": "Service Desk Use Only",
"order": 21
}
]
}
]
}Cause
When an existing setup of JSM DC and Confluence DataCenter is done it is connected via AppLink/tunnel at the same time. Now when a migration of Confluence DC is done, the AppLink is not removed automatically or disabled. Jira identifies one App-linked Confluence instance as PRIMARY. The searches on non-project-specific HelpCenter are performed exclusively against the primary Confluence even if more are linked and used, and that is by design.
Thus it creates a conflict in search where the Confluence Cloud doesn't require any AppLink integration.
In the past, you might have created integrations with other applications through Application Links. Although this would still work for Confluence Cloud, using it as the knowledge base requires that you create the integration through the Confluence Cloud integration page. No AppLinks this time.
On the Application Links settings page, you can define which of the connected Confluence instances should be the Primary instance. From now on, you can use the Confluence Cloud Integration to override the Application Links settings specifically for Knowledge Base functionality. This means that customers accessing the main Help Center page will only see Knowledge Base results from the Primary Confluence instance when searching. Additionally, project administrators will have this setting as the default option when configuring Knowledge Bases in Jira Service Management projects.
Solution
Workaround 1
Turn on the following feature flag: sd.kb.cloud.always.primary . This will make Confluence Cloud Integration override the PRIMARY Confluence App Link for KB search purposes only.
For Feature Flag instructions please go to: How to manage dark features in Jira Server and Data Center
Workaround 2
- Verify if there are any AppLink configured that are used to connect with the Confluence Cloud.
- If any AppLink with Confluence Cloud is identified, it is recommended to delete it.