Google Indexed our variation page

[#1]

Hello, we recently ran a split test for our homepage. The variation was called /index_new after the url which we didn't want indexing so we put a robotsnoindex on it but google has somehow indexed this page and affected our rankings because of duplicate content. Why has this happened?

[#2]

How did you put the "noindex" to that page? By adding a meta-tag to the head-section via jQuery?

[#3]

Where did you put the "noindex"? In the robots.txt file or in the document as a meta tag?
Anyway, for what I know it's a best practice to use the rel=canonical for this situation. Where you say that "index_new" belongs to "index", so the search engine would not consider it as a duplicate.