From a8ee4d543a693fe358862559902d6e65aa08fc4f Mon Sep 17 00:00:00 2001
From: Piotr Dulikowski <piodul@scylladb.com>
Date: Mon, 30 Oct 2023 16:05:11 +0100
Subject: [PATCH] raft: transfer information about group0 liveness in
 direct_fd_ping

Add a new variant of the reply to the direct_fd_ping which specifies
whether the local group0 is alive or not, and start actively using it.

There is no need to introduce a cluster feature. Due to how our
serialization framework works, nodes which do not recognize the new
variant will treat it as the existing std::monostate. The std::monostate
means "the node and group0 is alive"; nodes before the changes in this
commit would send a std::monostate anyway, so this is completely
transparent for the old nodes.
---
 idl/raft.idl.hh                     |  6 +++++-
 service/raft/group0_fwd.hh          |  6 +++++-
 service/raft/raft_group_registry.cc | 24 +++++++++++++++++++-----
 3 files changed, 29 insertions(+), 7 deletions(-)

diff --git a/idl/raft.idl.hh b/idl/raft.idl.hh
index 658a190812..81845257c5 100644
--- a/idl/raft.idl.hh
+++ b/idl/raft.idl.hh
@@ -121,8 +121,12 @@ struct wrong_destination {
     raft::server_id reached_id;
 };
 
+struct group_liveness_info {
+    bool group0_alive;
+};
+
 struct direct_fd_ping_reply {
-    std::variant<std::monostate, service::wrong_destination> result;
+    std::variant<std::monostate, service::wrong_destination, service::group_liveness_info> result;
 };
 
 verb [[with_client_info, cancellable]] direct_fd_ping (raft::server_id dst_id) -> service::direct_fd_ping_reply;
diff --git a/service/raft/group0_fwd.hh b/service/raft/group0_fwd.hh
index 815b86abaf..bc42a5c42e 100644
--- a/service/raft/group0_fwd.hh
+++ b/service/raft/group0_fwd.hh
@@ -75,8 +75,12 @@ struct wrong_destination {
     raft::server_id reached_id;
 };
 
+struct group_liveness_info {
+    bool group0_alive;
+};
+
 struct direct_fd_ping_reply {
-    std::variant<std::monostate, wrong_destination> result;
+    std::variant<std::monostate, wrong_destination, group_liveness_info> result;
 };
 
 } // namespace service
diff --git a/service/raft/raft_group_registry.cc b/service/raft/raft_group_registry.cc
index 4b3c9bcc51..3b7f4378f8 100644
--- a/service/raft/raft_group_registry.cc
+++ b/service/raft/raft_group_registry.cc
@@ -297,15 +297,27 @@ void raft_group_registry::init_rpc_verbs() {
         // XXX: update address map here as well?
 
         if (_my_id != dst) {
-            co_return direct_fd_ping_reply {
+            return make_ready_future<direct_fd_ping_reply>(direct_fd_ping_reply {
                 .result = wrong_destination {
                     .reached_id = _my_id,
                 },
-            };
+            });
         }
-        co_return direct_fd_ping_reply {
-            .result = std::monostate{},
-        };
+
+        return container().invoke_on(0, [] (raft_group_registry& me) -> future<direct_fd_ping_reply> {
+            bool group0_alive = false;
+            if (me._group0_id) {
+                auto* group0_server = me.find_server(*me._group0_id);
+                if (group0_server && group0_server->is_alive()) {
+                    group0_alive = true;
+                }
+            }
+            co_return direct_fd_ping_reply {
+                .result = service::group_liveness_info{
+                    .group0_alive = group0_alive,
+                }
+            };
+        });
     });
 }
 
@@ -513,6 +525,8 @@ future<bool> direct_fd_pinger::ping(direct_failure_detector::pinger::endpoint_id
             rslog.trace("ping(id = {}, ip_addr = {}): wrong destination (reached {})",
                         dst_id, *addr, wrong_dst->reached_id);
             co_return false;
+        } else if (auto* info = std::get_if<group_liveness_info>(&reply.result)) {
+            co_return info->group0_alive;
         }
     } catch (seastar::rpc::closed_error&) {
         co_return false;