diff --git a/swh/storage/in_memory.py b/swh/storage/in_memory.py
--- a/swh/storage/in_memory.py
+++ b/swh/storage/in_memory.py
@@ -75,13 +75,15 @@
 
     def _content_add(self, contents, with_data):
         if self.journal_writer:
-            for content in contents:
-                content = attr.evolve(content, data=None)
+            for (content, origin) in contents:
+                content = attr.evolve(content, data=None).to_dict()
+                if origin is not None:
+                    content['origin'] = origin
                 self.journal_writer.write_addition('content', content)
 
         content_with_data = []
         content_without_data = []
-        for content in contents:
+        for (content, origin) in contents:
             if content.status is None:
                 content.status = 'visible'
             if content.length is None:
@@ -89,7 +91,7 @@
             if content.status == 'visible':
                 content_with_data.append(content)
             elif content.status == 'absent':
-                content_without_data.append(content)
+                content_without_data.append((content, origin))
 
         count_content_added, count_content_bytes_added = \
             self._content_add_present(content_with_data, with_data)
@@ -139,17 +141,28 @@
 
     def _content_add_absent(self, contents):
         count = 0
+        if not contents:
+            # Shortcut, needed for the zip() below to not return []
+            return count
+        (contents, origins) = zip(*contents)
         skipped_content_missing = self.skipped_content_missing(contents)
-        for content in skipped_content_missing:
+        for (content, origin) in zip(skipped_content_missing, origins):
             key = self._content_key(content)
             for algo in DEFAULT_ALGORITHMS:
                 self._skipped_content_indexes[algo][content.get_hash(algo)] \
                     .add(key)
-            self._skipped_contents[key] = content
+            self._skipped_contents[key] = (content, origin)
             count += 1
 
         return count
 
+    def _content_to_model(self, contents):
+        """Takes a list of content dicts, optionally with an extra 'origin'
+        key, and returns a list of tuples (model.Content, origin)."""
+        contents = (cont.copy() for cont in contents)
+        contents = ((cont, cont.pop('origin', None)) for cont in contents)
+        return [(Content.from_dict(cont), orig) for (cont, orig) in contents]
+
     def content_add(self, content):
         """Add content blobs to the storage
 
@@ -179,9 +192,9 @@
                 skipped_content:add: New skipped contents (no data) added
 
         """
-        content = [Content.from_dict(c) for c in content]
+        content = self._content_to_model(content)
         now = datetime.datetime.now(tz=datetime.timezone.utc)
-        for item in content:
+        for (item, _) in content:
             item.ctime = now
         return self._content_add(content, with_data=True)
 
@@ -214,7 +227,7 @@
                 skipped_content:add: New skipped contents (no data) added
 
         """
-        content = [Content.from_dict(c) for c in content]
+        content = self._content_to_model(content)
         return self._content_add(content, with_data=False)
 
     def content_get(self, content):
diff --git a/swh/storage/storage.py b/swh/storage/storage.py
--- a/swh/storage/storage.py
+++ b/swh/storage/storage.py
@@ -198,6 +198,15 @@
                     raise
 
         if content_without_data:
+            content_without_data = \
+                [cont.copy() for cont in content_without_data]
+            origins = db.origin_get_by_url(
+                [cont.get('origin') for cont in content_without_data],
+                cur=cur)
+            for (cont, origin) in zip(content_without_data, origins):
+                origin = dict(zip(db.origin_cols, origin))
+                if 'origin' in cont:
+                    cont['origin'] = origin['id']
             db.mktemp('skipped_content', cur)
             db.copy_to(content_without_data, 'tmp_skipped_content',
                        db.skipped_content_keys, cur)
diff --git a/swh/storage/tests/test_storage.py b/swh/storage/tests/test_storage.py
--- a/swh/storage/tests/test_storage.py
+++ b/swh/storage/tests/test_storage.py
@@ -119,6 +119,7 @@
                 '2c772cc193778aac9a137b8dc5834b9b'),
             'reason': 'Content too long',
             'status': 'absent',
+            'origin': 'file:///dev/zero',
         }
 
         self.skipped_cont2 = {