Revert "fix: slow responses on schema cache reload"

This reverts commit 727ef465c1.

Also documents requests waiting for the schema cache.

CHANGELOG.md

@@ -24,7 +24,6 @@ This project adheres to [Semantic Versioning](http://semver.org/).
  - #3237, Dump media handlers and timezones with --dump-schema - @wolfgangwalther
  - #3323, #3324, Don't hide error on LISTEN channel failure - @steve-chavez
  - #3330, Incorrect admin server `/ready` response on slow schema cache loads - @steve-chavez
- - #3327, Fix slow responses on schema cache reloads - @steve-chavez
  - #3340, Log when the LISTEN channel gets a notification - @steve-chavez
  - #3345, Fix in-database configuration values not loading for `pgrst.server_trace_header` and `pgrst.server_cors_allowed_origins` - @laurenceisla
  - #3361, Clarify PGRST204(column not found) error message - @steve-chavez

docs/references/schema_cache.rst

@@ -18,8 +18,8 @@ You can do this with UNIX signals or with PostgreSQL notifications. It's also po
 
 .. note::
 
+  - Requests will wait until the schema cache reload is done. This to prevent client errors due to an stale schema cache.
   - If you are using the :ref:`in_db_config`, a schema cache reload will :ref:`reload the configuration<config_reloading>` as well.
-  - There’s no downtime when reloading the schema cache. The reloading will happen on a background thread while serving requests.
 
 .. _schema_reloading_signals:
 

src/PostgREST/AppState.hs

@@ -320,12 +320,10 @@ loadSchemaCache appState observer = do
           return SCOnRetry
 
     Right sCache -> do
+      putSchemaCache appState $ Just sCache
       observer $ SchemaCacheQueriedObs resultTime
       (t, _) <- timeItT $ observer $ SchemaCacheSummaryObs sCache
       observer $ SchemaCacheLoadedObs t
-      -- it's important to update AppState schema cache only once it has been fully evaluated before (this will be done by the observer above)
-      -- otherwise requests will wait for the schema cache to be loaded
-      putSchemaCache appState $ Just sCache
       putSchemaCacheLoaded appState True
       return SCLoaded
 

test/io/test_big_schema.py

@@ -7,28 +7,8 @@ from util import *
 from postgrest import *
 
 
-def test_first_request_succeeds(defaultenv):
+def test_requests__wait_for_schema_cache_reload(defaultenv):
-    "the first request suceeds fast since the admin server readiness was checked before"
+    "requests that use the schema cache (e.g. resource embedding) wait for the schema cache to reload"
-
-    env = {
-        **defaultenv,
-        "PGRST_DB_SCHEMAS": "apflora",
-        "PGRST_DB_POOL": "2",
-        "PGRST_DB_ANON_ROLE": "postgrest_test_anonymous",
-        "PGRST_SERVER_TIMING_ENABLED": "true",
-    }
-
-    with run(env=env, wait_max_seconds=30) as postgrest:
-        response = postgrest.session.get("/tpopmassn?select=*,tpop(*)")
-        assert response.status_code == 200
-
-        server_timings = parse_server_timings_header(response.headers["Server-Timing"])
-        plan_dur = server_timings["plan"]
-        assert plan_dur < 2.0
-
-
-def test_requests_do_not_wait_for_schema_cache_reload(defaultenv):
-    "requests that use the schema cache (e.g. resource embedding) do not wait for the schema cache to reload"
 
     env = {
         **defaultenv,
@@ -48,10 +28,9 @@ def test_requests_do_not_wait_for_schema_cache_reload(defaultenv):
         response = postgrest.session.get("/tpopmassn?select=*,tpop(*)")
         assert response.status_code == 200
 
-        # response should be fast
         server_timings = parse_server_timings_header(response.headers["Server-Timing"])
         plan_dur = server_timings["plan"]
-        assert plan_dur < 2.0
+        assert plan_dur > 10000.0
 
 
 # TODO: This test fails now because of https://github.com/PostgREST/postgrest/pull/2122