There has been an apparent case where the question poster asked a question that had been answered in their own paper from about 10 years before the question. This would not be a problem if the question poster had answered the question themself; however, they did not. The fact that the question poster had a paper with the solution before the question was discovered recently. The discoverer brought up the topic of "reputation farming"; however, I personally think the more relevant concern was brought up in another comment: that posing a question with a known answer without posting that answer abuses the time of the answerer (and as such, might be considered rep farming because it gains reputation in a dubious manner).
Is there a policy on this, and should there be one? This seems like a circumstance that would happen very rarely - even more so because many posters don't associate their name with their MO account, and so uncovering it would be difficult.