Include node friendly names in runner error messages

Each error in the combined message is now prefixed with the node's friendly
name (e.g. "MacBook Pro: OOM; Mac Studio: connection reset") so the root
cause node is easily identifiable.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

dashboard/src/routes/+page.svelte

@@ -1278,7 +1278,7 @@
           typeof runnerData === "object"
         ) {
           const rd = runnerData as { errorMessage?: string };
-          if (rd.errorMessage) failedErrors.push(rd.errorMessage);
+          if (rd.errorMessage) failedErrors.push(`${getNodeName(nodeId)}: ${rd.errorMessage}`);
         }
       }
       if (status) {

src/exo/master/process_managers/instance_health.py

@@ -23,7 +23,7 @@ class InstanceHealthReconciler:
                 continue
 
             is_failed, error_message = instance_runners_failed(
-                instance, state.runners
+                instance, state.runners, state.node_identities
             )
             if is_failed:
                 events.append(

src/exo/master/reconcile.py

@@ -10,7 +10,7 @@ from exo.shared.types.commands import PlaceInstance
 from exo.shared.types.common import MetaInstanceId, NodeId
 from exo.shared.types.events import Event
 from exo.shared.types.meta_instance import MetaInstance
-from exo.shared.types.profiling import MemoryUsage, NodeNetworkInfo
+from exo.shared.types.profiling import MemoryUsage, NodeIdentity, NodeNetworkInfo
 from exo.shared.types.topology import RDMAConnection, SocketConnection
 from exo.shared.types.worker.instances import (
     BaseInstance,
@@ -100,6 +100,7 @@ def instance_connections_healthy(instance: Instance, topology: Topology) -> bool
 def instance_runners_failed(
     instance: Instance,
     runners: Mapping[RunnerId, RunnerStatus],
+    node_identities: Mapping[NodeId, NodeIdentity],
 ) -> tuple[bool, str | None]:
     """Check if an instance's runners have all reached terminal failure states.
 
@@ -114,6 +115,12 @@ def instance_runners_failed(
     if not instance_runner_ids:
         return False, None
 
+    # Build reverse mapping: runner_id -> node_id
+    runner_to_node: dict[RunnerId, NodeId] = {
+        runner_id: node_id
+        for node_id, runner_id in instance.shard_assignments.node_to_runner.items()
+    }
+
     has_any_failed = False
     error_messages: list[str] = []
 
@@ -125,7 +132,9 @@ def instance_runners_failed(
         if isinstance(status, RunnerFailed):
             has_any_failed = True
             if status.error_message:
-                error_messages.append(status.error_message)
+                node_id = runner_to_node.get(runner_id)
+                name = node_identities[node_id].friendly_name if node_id and node_id in node_identities else node_id or "unknown"
+                error_messages.append(f"{name}: {status.error_message}")
         elif isinstance(status, RunnerShutdown):
             pass  # Terminal but not a failure indicator on its own
         else:

src/exo/master/tests/test_reconcile.py

@@ -472,7 +472,7 @@ def test_runners_failed_all_failed():
         rid: RunnerFailed(error_message="OOM")
         for rid in inst.shard_assignments.node_to_runner.values()
     }
-    is_failed, error = instance_runners_failed(inst, runners)
+    is_failed, error = instance_runners_failed(inst, runners, {})
     assert is_failed is True
     assert error is not None
     assert "OOM" in error
@@ -486,9 +486,10 @@ def test_runners_failed_mixed_failed_shutdown():
         runner_ids[0]: RunnerFailed(error_message="crash"),
         runner_ids[1]: RunnerShutdown(),
     }
-    is_failed, error = instance_runners_failed(inst, runners)
+    is_failed, error = instance_runners_failed(inst, runners, {})
     assert is_failed is True
-    assert error == "crash"
+    assert error is not None
+    assert "crash" in error
 
 
 def test_runners_not_failed_all_shutdown():
@@ -498,7 +499,7 @@ def test_runners_not_failed_all_shutdown():
         rid: RunnerShutdown()
         for rid in inst.shard_assignments.node_to_runner.values()
     }
-    is_failed, _ = instance_runners_failed(inst, runners)
+    is_failed, _ = instance_runners_failed(inst, runners, {})
     assert is_failed is False
 
 
@@ -510,14 +511,14 @@ def test_runners_not_failed_still_active():
         runner_ids[0]: RunnerFailed(error_message="OOM"),
         runner_ids[1]: RunnerLoading(),
     }
-    is_failed, _ = instance_runners_failed(inst, runners)
+    is_failed, _ = instance_runners_failed(inst, runners, {})
     assert is_failed is False
 
 
 def test_runners_not_failed_no_status():
     """Runner not yet reported = not failed."""
     _, inst = _instance(node_ids=["node-a"])
-    is_failed, _ = instance_runners_failed(inst, {})
+    is_failed, _ = instance_runners_failed(inst, {}, {})
     assert is_failed is False
 
 
@@ -528,7 +529,7 @@ def test_runners_not_failed_healthy():
         rid: RunnerReady()
         for rid in inst.shard_assignments.node_to_runner.values()
     }
-    is_failed, _ = instance_runners_failed(inst, runners)
+    is_failed, _ = instance_runners_failed(inst, runners, {})
     assert is_failed is False