It "succeeded" in that it delivers on its promise and some people still use it, but it is largely dead now due to the lack of development and the community's inability to push the language itself forward. There are forks that address that, but Elm itself seems unlikely to Lazarus any time soon.
It's definitely proof that software can be written in such a regime, though, and I hope we see something similarly dogmatic some day.
Gleam / Lustre are fantastic, and I hope to use them in anger at some point, but it's my understanding that they don't aim to be as "total" as Elm (quotation marks because I don't think Elm was total in the formal sense). That is, while Lustre is very much following in Elm's footsteps, both it and the language have not been designed around avoiding runtime failure at all costs. (Which makes sense, given the Erlang/BEAM heritage.)
It's definitely proof that software can be written in such a regime, though, and I hope we see something similarly dogmatic some day.