diff --git a/tidal-server/src/cluster/node.rs b/tidal-server/src/cluster/node.rs
index 0db2216..e10f68e 100644
--- a/tidal-server/src/cluster/node.rs
+++ b/tidal-server/src/cluster/node.rs
@@ -3165,6 +3165,26 @@ impl ShardReplica {
             return;
         }
         self.self_restart_refused.store(false, Ordering::Release);
+        // m12 reseed-loop-fix: re-check the marker before committing to the exit.
+        // A snapshot_required latch from a TRANSIENT classification — the leader's
+        // baseline advanced past this node's persisted frontier while it was down,
+        // so the first heartbeat's decide_join saw it briefly behind — self-heals
+        // via the stream: `clear_stale_reseed_marker_if_caught_up` clears the marker
+        // once caught up, journaling the durable term marker on the clean join. The
+        // quorum poll above took time (a blocking peer fan-out); if the marker
+        // healed meanwhile, a reboot would reseed NOTHING (the leader answers
+        // needed=false for a caught-up shard), so self-restarting is futile and
+        // loops. Abort. Only a marker that CANNOT self-heal (a genuine compacted
+        // gap, still latched here) proceeds to the restart.
+        if !self.reseed_marker_latched.load(Ordering::Acquire) {
+            tracing::info!(
+                region = %self.region_name,
+                "reseed_self_restart: the reseed marker healed via stream catch-up during the \
+                 quorum check — the shard is caught up, a reboot would reseed nothing. Aborting \
+                 the self-restart (m12 reseed-loop-fix)."
+            );
+            return;
+        }
         // m12 reseed-loop-fix (Fix 3): the process-wide exit is owned by the
         // node-level coordinator, not this single shard. It fires once, only
         // after every hosted shard has also requested a restart or a bounded
@@ -3199,6 +3219,21 @@ impl ShardReplica {
                         if deadline > now {
                             std::thread::sleep(deadline - now);
                         }
+                        // m12 reseed-loop-fix: the grace window is exactly when a
+                        // transient snapshot_required latch self-heals via stream
+                        // catch-up. Re-check before firing — if THIS shard's marker
+                        // healed, a reboot would reseed nothing (caught up), so abort
+                        // the futile self-restart. Only a still-latched marker (a
+                        // genuine compacted gap) fires the exit.
+                        if !node.reseed_marker_latched.load(Ordering::Acquire) {
+                            tracing::info!(
+                                region = %node.region_name,
+                                "reseed_self_restart: the reseed marker healed via stream \
+                                 catch-up during the Fix 3 grace — aborting the self-restart \
+                                 (the shard is caught up; a reboot would reseed nothing)."
+                            );
+                            return;
+                        }
                         if super::reseed_restart::fire_due() {
                             node.fire_graceful_self_restart();
                         }