I’m in the process of generating Staging environments for an application linked Confluence Server and Jira Software Server. A few hundred pages are heavily linked with Jira issue references, which usually create a “<#> Jira Links” pill at the top of the confluence page.
After recreating the app link with new Server IDs, I now have pages which the page links work perfectly and are correctly reciprocated in our staging Jira, but the summaries have disappeared from the top of the pages.
I’m trying to figure out where in the confluence database, if at all, the code which generates that “<#> Jira Links” pill is stored for these pages to correct them; I’m concerned about it creating a false-positive/negative in my XML export testing. 
TIA for anyone who can help shed some light on this!