From 301935f8c1f2ad48c0b5fb1667727a0bae747926 Mon Sep 17 00:00:00 2001
From: Graham McIntire <graham@mcintire.me>
Date: Mon, 27 Apr 2026 14:22:10 -0500
Subject: [PATCH] fix: log swallowed async task exits in GEFS scoring and HRRR
 range fetch

GefsFetch silently dropped crashed score-computation tasks via
`{:exit, _reason} -> []`, hiding partial-grid outages from k8s logs.
HRRR range downloads mapped exits to `{:error, reason}` without logging
the URL, so the failing range was lost by the time the caller surfaced
the error.

All three sites now Logger.error with enough context (fh, valid_time,
url) to debug from production logs.
---
 lib/microwaveprop/weather/hrrr_client.ex       | 18 ++++++++++++++----
 lib/microwaveprop/workers/gefs_fetch_worker.ex | 11 +++++++++--
 2 files changed, 23 insertions(+), 6 deletions(-)

diff --git a/lib/microwaveprop/weather/hrrr_client.ex b/lib/microwaveprop/weather/hrrr_client.ex
index 20277302..0e0190b6 100644
--- a/lib/microwaveprop/weather/hrrr_client.ex
+++ b/lib/microwaveprop/weather/hrrr_client.ex
@@ -566,8 +566,13 @@ defmodule Microwaveprop.Weather.HrrrClient do
         timeout: 120_000
       )
       |> Enum.map(fn
-        {:ok, result} -> result
-        {:exit, reason} -> {:error, reason}
+        {:ok, result} ->
+          result
+
+        {:exit, reason} ->
+          Logger.error("HRRR grib range download crashed (to_file): url=#{url} reason=#{inspect(reason)}")
+
+          {:error, reason}
       end)
 
     case Enum.find(results, &match?({:error, _}, &1)) do
@@ -612,8 +617,13 @@ defmodule Microwaveprop.Weather.HrrrClient do
         timeout: 120_000
       )
       |> Enum.map(fn
-        {:ok, result} -> result
-        {:exit, reason} -> {:error, reason}
+        {:ok, result} ->
+          result
+
+        {:exit, reason} ->
+          Logger.error("HRRR grib range download crashed: url=#{url} reason=#{inspect(reason)}")
+
+          {:error, reason}
       end)
 
     # Sort by offset and concatenate
diff --git a/lib/microwaveprop/workers/gefs_fetch_worker.ex b/lib/microwaveprop/workers/gefs_fetch_worker.ex
index a998426d..0cc13c4b 100644
--- a/lib/microwaveprop/workers/gefs_fetch_worker.ex
+++ b/lib/microwaveprop/workers/gefs_fetch_worker.ex
@@ -172,8 +172,15 @@ defmodule Microwaveprop.Workers.GefsFetchWorker do
         timeout: 30_000
       )
       |> Stream.flat_map(fn
-        {:ok, results} -> results
-        {:exit, _reason} -> []
+        {:ok, results} ->
+          results
+
+        {:exit, reason} ->
+          Logger.error(
+            "GefsFetch scoring task crashed: fh=#{fh} valid_time=#{inspect(valid_time)} reason=#{inspect(reason)}"
+          )
+
+          []
       end)
       |> Enum.to_list()