diff --git a/CHANGELOG.md b/CHANGELOG.md
index b26195320f..cb8d1c8189 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -1,5 +1,11 @@
 # Changelog
 
+## Unreleased
+
+**Bug Fixes**:
+
+- Accept incoming requests even if there was an error fetching their project config. ([#4140](https://github.com/getsentry/relay/pull/4140))
+
 ## 24.11.1
 
 **Breaking Changes**:
diff --git a/relay-server/src/services/projects/cache/service.rs b/relay-server/src/services/projects/cache/service.rs
index 435355c106..01463b1970 100644
--- a/relay-server/src/services/projects/cache/service.rs
+++ b/relay-server/src/services/projects/cache/service.rs
@@ -124,9 +124,7 @@ impl ProjectCacheService {
                         "failed to fetch project from source: {fetch:?}"
                     );
 
-                    // TODO: change this to ProjectState::Pending once we consider it safe to do so.
-                    // see https://github.com/getsentry/relay/pull/4140.
-                    ProjectState::Disabled.into()
+                    ProjectState::Pending.into()
                 }
             };
 
diff --git a/tests/integration/test_query.py b/tests/integration/test_query.py
index 6f05f5e9f4..4ee73aa328 100644
--- a/tests/integration/test_query.py
+++ b/tests/integration/test_query.py
@@ -155,24 +155,18 @@ def get_project_config():
         mini_sentry.clear_test_failures()
 
 
-def test_query_retry_maxed_out(mini_sentry, relay_with_processing, events_consumer):
+def test_query_retry_maxed_out(mini_sentry, relay):
     """
     Assert that a query is not retried an infinite amount of times.
-
-    This is not specific to processing or store, but here we have the outcomes
-    consumer which we can use to assert that an event has been dropped.
     """
     request_count = 0
 
-    events_consumer = events_consumer()
-
     original_get_project_config = mini_sentry.app.view_functions["get_project_config"]
 
     @mini_sentry.app.endpoint("get_project_config")
     def get_project_config():
         if flask_request.json.get("global") is True:
             return original_get_project_config()
-
         nonlocal request_count
         request_count += 1
         print("RETRY", request_count)
@@ -180,14 +174,11 @@ def get_project_config():
 
     RETRIES = 1
     query_timeout = 0.5  # Initial grace period
-
     # Relay's exponential backoff: INITIAL_INTERVAL = 1s; DEFAULT_MULTIPLIER = 1.5;
     for retry in range(RETRIES):  # 1 retry
         query_timeout += 1 * 1.5 ** (retry + 1)
 
-    relay = relay_with_processing(
-        {"limits": {"query_timeout": math.ceil(query_timeout)}}
-    )
+    relay = relay(mini_sentry, {"limits": {"query_timeout": math.ceil(query_timeout)}})
 
     # No error messages yet
     assert mini_sentry.test_failures.empty()
@@ -195,11 +186,16 @@ def get_project_config():
     try:
         relay.send_event(42)
         time.sleep(query_timeout)
-
         assert request_count == 1 + RETRIES
         assert {str(e) for _, e in mini_sentry.current_test_failures()} == {
             "Relay sent us event: error fetching project states: upstream request returned error 500 Internal Server Error: no error details",
         }
+
+        time.sleep(1)  # Wait for project to be cached
+
+        # Relay still accepts events for this project
+        next_response = relay.send_event(42)
+        assert "id" in next_response
     finally:
         mini_sentry.clear_test_failures()