@cheatfate I don't think I follow what you meant. The links you sent don't match with what you mention (I believe). There is a TransportUseClosedError handled (and re-reased) nearby, but none of those links points to it.

One of them points to this piece of code in the same function as the aforementioned except TransportUseClosedError:

  except HttpError as exc:
    debug "Http Error", exc = exc.msg
  except AsyncStreamError as exc:

And the other points to the parameters of the next function, which seems to be unrelated to any TransportUseClosedError:

  hostname: string,
  address: MultiAddress,
  peerId: Opt[PeerId] = Opt.none(PeerId)): Future[Connection] {.async.} =

Do you mean that the "normal shutdown process" somehow triggers the except CatchableError line in the accept proc in switch.nim?

Everything what falls into

error "Exception in accept loop, exiting", exc = exc.msg

looks like serious error, and nimbus consumers actually asking what the error is. But this error happens when TransportUseClosedError happens and got catched by this handler. This specific exception MUST not be reported as error type of logs.

Please, next time it'd be great to have what the expected behavior should be when describing the issue.

@cheatfate do you have more logs related to it? The error msg "Transport closed, no more connections!" is created by

Was there a log "Server was closed" before the one you reported? If it was indeed caused by TransportUseClosedError, there should be one