Issue Summary

It is not possible to search for Confluence Cloud spaces when linking Confluence page from Jira issue. User get an error "Unable to collect information from the selected Confluence instance".

Environment

• Jira Server, application link to Confluence Cloud

Steps to Reproduce

1. Link Confluence Cloud instance to Jira Server
2. Go to an Issue > Link > Confluence page > Search for a page

Expected Results

The space search is performed and results are returned.

Actual Results

The following error is thrown in the browser:

"Unable to collect information from the selected Confluence instance"

Enabling org.apache.http log reveals that Jira is hitting Confluence Cloud endpoint that does not exist:

2019-08-29 11:06:54,678 http-nio-8080-exec-1595 DEBUG admin 666x744009x1 hkkf9o  xxx.xxx.xxx.xxx,yyy.yyy.yyy.yyy /rest/confluenceIssueLink/1/confluence/space [o.apache.http.wire] http-outgoing-10263 >> "POST /wiki/rpc/xmlrpc HTTP/1.1[\r][\n]"
...
http-outgoing-10263 << "HTTP/1.1 404 Not Found[\r][\n]"
...
2019-08-29 11:06:54,713 http-nio-8080-exec-1595 DEBUG admin 666x744009x1 hkkf9o xxx.xxx.xxx.xxx,yyy.yyy.yyy.yyy /rest/confluenceIssueLink/1/confluence/space [o.apache.http.wire] http-outgoing-10263 << "<h1>404 Not Found</h1><p>Oops, you've found a dead link. Click here to return to the <a href="javascript:window.history.back()">previous page</a></p>"

Notes

The actual cause for this bug is SOAP / XML-RPC deprecation on Confluence: https://blog.developer.atlassian.com/confluence-cloud-soap-api-migration/. However, due to the fact that the endpoint has been preserved on Confluence Server, we cannot reproduce that issue if Jira is linked to Confluence Server.

The page search method should be replaced with GET /wiki/rest/api/space.

Workaround

Link a Confluence page directly to Jira issue.