Ask HN: Has anyone had issues with RDS failing over?

Recently I've seen multiple instances of RDS failing over, across many AWS accounts in a few regions.

Something that started happening sporadically while services were running fine for a year at at time before that.

Typically involves Freeable memory going to zero and SWAP disappearing completely, as if it is unmounted from the instance.

Support has been cryptic. Accounts are not connected in any way whatsoever. Have any of you experienced that?