On ticket searches for example /p/test/tickets/search/?q=foo we have these references which are wrong:
<link rel="alternate" type="application/rss+xml" title="RSS" href="feed.rss"/>
<link rel="alternate" type="application/atom+xml" title="Atom" href="feed.atom"/>
Check the ForgeTracker code to see if we have any support for feeds of a ticket search. I think probably not, in which case we should remove those links. (Make sure to keep them on other types of ticket pages where appropriate)
And then ideally we should make those urls like ../search/feed.rss and ../search/feed.atom return a 410 Gone status so bots stop trying to crawl them.