It seems that JIRA recently complains about collation, which it would be a good thing unless it would complain about a valid collation being as incorrect.

We are using a database with UTF-8 encoding and C.UTF-8 collation, which is it the correct collation to use for an UTF-8 database.

****************************************************************************************************
You are using an unsupported postgres72 collation: C.UTF-8. This may cause some functionality to not work.
Please use C or POSIX as the collation instead.
****************************************************************************************************

Having a database with encoding "C" would be Unicode compatible and will clearly crease serious issues when dealing with text data.

It seems that you must consider C.UTF-8 and POSIX.UTF-8 as valid encodings. In fact this warning message should include a link to a Confluence page that documents the correct database encodings to use.