diff --git a/brozzler/worker.py b/brozzler/worker.py
index 1aef258..1c65036 100644
--- a/brozzler/worker.py
+++ b/brozzler/worker.py
@@ -205,6 +205,8 @@ class BrozzlerWorker:
                     'got "%s %s" response on warcprox '
                     'WARCPROX_WRITE_RECORD request (expected 204)',
                     e.getcode(), e.info())
+        except urllib.error.URLError as e:
+            raise brozzler.ProxyError('_warcprox_write_record: %s', e)
 
     def _remember_videos(self, page, ydl_spy):
         if not 'videos' in page:
diff --git a/setup.py b/setup.py
index fb5b62a..86c72c6 100644
--- a/setup.py
+++ b/setup.py
@@ -32,7 +32,7 @@ def find_package_data(package):
 
 setuptools.setup(
         name='brozzler',
-        version='1.1b11.dev234',
+        version='1.1b11.dev235',
         description='Distributed web crawling with browsers',
         url='https://github.com/internetarchive/brozzler',
         author='Noah Levitt',
diff --git a/tests/test_units.py b/tests/test_units.py
index d9825d5..911fceb 100644
--- a/tests/test_units.py
+++ b/tests/test_units.py
@@ -146,6 +146,15 @@ def test_proxy_down():
         with pytest.raises(brozzler.ProxyError):
             worker._fetch_url(site, page)
 
+        # WARCPROX_WRITE_RECORD
+        with pytest.raises(brozzler.ProxyError):
+            worker._warcprox_write_record(
+                    warcprox_address=not_listening_proxy,
+                    url='test://proxy_down/warcprox_write_record',
+                    warc_type='metadata',
+                    content_type='text/plain',
+                    payload=b'''payload doesn't matter here''')
+
 def test_start_stop_backwards_compat():
     site = brozzler.Site(None, {'seed': 'http://example.com/'})
     assert len(site.starts_and_stops) == 1