fixup! approval-distribution: Update topology if authorities are disc…

…overed later (#2981)

The previous fix was actually incomplete because we update the
authorties only on the situation where we decided to reconnect because
we had a low connectivity issue. Now the problem is that
update_authority_ids use the list of connected peers, so on restart that
does contain anything, so calling immediately after
issue_connection_request won't detect all authorties, so we need to also
check every block as the comment said, but that did not match the code.

polkadot/node/network/gossip-support/src/lib.rs

@@ -271,8 +271,30 @@ where
 					)
 					.await?;
 				}
-				// authority_discovery is just a cache so let's try every leaf to detect if there
-				// are new authorities there.
+				// authority_discovery is just a cache so let's try every time we try to re-connect
+				// if new authorities are present.
+				self.update_authority_ids(sender, session_info.discovery_keys).await;
+			} else if let Some(session_index) = self
+				.last_session_index
+				.filter(|last_session_index| *last_session_index >= current_index)
+			{
+				// authority_discovery is just a cache so let's try every leaf from the current
+				// session if there are new authorities detected and inform the needed subsystems.
+				let session_info =
+					util::request_session_info(leaf, session_index, sender).await.await??;
+
+				let session_info = match session_info {
+					Some(s) => s,
+					None => {
+						gum::warn!(
+							relay_parent = ?leaf,
+							session_index = self.last_session_index,
+							"Failed to get session info.",
+						);
+
+						continue
+					},
+				};
 				self.update_authority_ids(sender, session_info.discovery_keys).await;
 			}
 		}