RUST-2154 Update tests that use WriteConcernFailed code name

src/test/spec/json/retryable-writes/unified/insertOne-serverErrors.json

@@ -739,7 +739,7 @@
       ]
     },
     {
-      "description": "InsertOne fails after WriteConcernError WriteConcernFailed",
+      "description": "InsertOne fails after WriteConcernError WriteConcernTimeout",
       "operations": [
         {
           "name": "failPoint",
@@ -757,7 +757,6 @@
                 ],
                 "writeConcernError": {
                   "code": 64,
-                  "codeName": "WriteConcernFailed",
                   "errmsg": "waiting for replication timed out",
                   "errInfo": {
                     "wtimeout": true

src/test/spec/json/retryable-writes/unified/insertOne-serverErrors.yml

@@ -339,7 +339,7 @@ tests:
           - { _id: 2, x: 22 }
           - { _id: 3, x: 33 } # The write was still applied.
   -
-    description: 'InsertOne fails after WriteConcernError WriteConcernFailed'
+    description: 'InsertOne fails after WriteConcernError WriteConcernTimeout'
     operations:
       -
         name: failPoint
@@ -353,7 +353,6 @@ tests:
               failCommands: [ insert ]
               writeConcernError:
                 code: 64
-                codeName: WriteConcernFailed
                 errmsg: 'waiting for replication timed out'
                 errInfo:
                   wtimeout: true

src/test/spec/json/transactions-convenient-api/README.md

@@ -0,0 +1,48 @@
+# Convenient API for Transactions Tests
+
+______________________________________________________________________
+
+## Introduction
+
+The YAML and JSON files in this directory are platform-independent tests meant to exercise a driver's implementation of
+the Convenient API for Transactions spec. These tests utilize the
+[Unified Test Format](../../unified-test-format/unified-test-format.md).
+
+Several prose tests, which are not easily expressed in YAML, are also presented in this file. Those tests will need to
+be manually implemented by each driver.
+
+## Prose Tests
+
+### Callback Raises a Custom Error
+
+Write a callback that raises a custom exception or error that does not include either UnknownTransactionCommitResult or
+TransientTransactionError error labels. Execute this callback using `withTransaction` and assert that the callback's
+error bypasses any retry logic within `withTransaction` and is propagated to the caller of `withTransaction`.
+
+### Callback Returns a Value
+
+Write a callback that returns a custom value (e.g. boolean, string, object). Execute this callback using
+`withTransaction` and assert that the callback's return value is propagated to the caller of `withTransaction`.
+
+### Retry Timeout is Enforced
+
+Drivers should test that `withTransaction` enforces a non-configurable timeout before retrying both commits and entire
+transactions. Specifically, three cases should be checked:
+
+- If the callback raises an error with the TransientTransactionError label and the retry timeout has been exceeded,
+    `withTransaction` should propagate the error to its caller.
+- If committing raises an error with the UnknownTransactionCommitResult label, and the retry timeout has been exceeded,
+    `withTransaction` should propagate the error to its caller.
+- If committing raises an error with the TransientTransactionError label and the retry timeout has been exceeded,
+    `withTransaction` should propagate the error to its caller. This case may occur if the commit was internally retried
+    against a new primary after a failover and the second primary returned a NoSuchTransaction error response.
+
+If possible, drivers should implement these tests without requiring the test runner to block for the full duration of
+the retry timeout. This might be done by internally modifying the timeout value used by `withTransaction` with some
+private API or using a mock timer.
+
+## Changelog
+
+- 2024-09-06: Migrated from reStructuredText to Markdown.
+- 2024-02-08: Converted legacy tests to unified format.
+- 2021-04-29: Remove text about write concern timeouts from prose test.

src/test/spec/json/transactions-convenient-api/unified/commit-writeconcernerror.json

@@ -56,7 +56,7 @@
   ],
   "tests": [
     {
-      "description": "commitTransaction is retried after WriteConcernFailed timeout error",
+      "description": "commitTransaction is retried after WriteConcernTimeout timeout error",
       "operations": [
         {
           "name": "failPoint",
@@ -74,7 +74,6 @@
                 ],
                 "writeConcernError": {
                   "code": 64,
-                  "codeName": "WriteConcernFailed",
                   "errmsg": "waiting for replication timed out",
                   "errInfo": {
                     "wtimeout": true
@@ -236,7 +235,7 @@
       ]
     },
     {
-      "description": "commitTransaction is retried after WriteConcernFailed non-timeout error",
+      "description": "commitTransaction is retried after WriteConcernTimeout non-timeout error",
       "operations": [
         {
           "name": "failPoint",
@@ -254,7 +253,6 @@
                 ],
                 "writeConcernError": {
                   "code": 64,
-                  "codeName": "WriteConcernFailed",
                   "errmsg": "multiple errors reported"
                 }
               }

src/test/spec/json/transactions-convenient-api/unified/commit-writeconcernerror.yml

@@ -32,7 +32,7 @@ initialData:
 
 tests:
   -
-    description: commitTransaction is retried after WriteConcernFailed timeout error
+    description: commitTransaction is retried after WriteConcernTimeout timeout error
     operations:
       - name: failPoint
         object: testRunner
@@ -47,7 +47,6 @@ tests:
               # with writeConcernError (see: SERVER-39292)
               writeConcernError:
                 code: 64
-                codeName: WriteConcernFailed
                 errmsg: "waiting for replication timed out"
                 errInfo: { wtimeout: true }
       - &operation
@@ -126,10 +125,10 @@ tests:
           - { _id: 1 }
   -
     # This test configures the fail point to return an error with the
-    # WriteConcernFailed code but without errInfo that would identify it as a
+    # WriteConcernTimeout code but without errInfo that would identify it as a
     # wtimeout error. This tests that drivers do not assume that all
-    # WriteConcernFailed errors are due to a replication timeout.
-    description: commitTransaction is retried after WriteConcernFailed non-timeout error
+    # WriteConcernTimeout errors are due to a replication timeout.
+    description: commitTransaction is retried after WriteConcernTimeout non-timeout error
     operations:
       - name: failPoint
         object: testRunner
@@ -144,7 +143,6 @@ tests:
               # with writeConcernError (see: SERVER-39292)
               writeConcernError:
                 code: 64
-                codeName: WriteConcernFailed
                 errmsg: "multiple errors reported"
       - *operation
     expectEvents: *expectEvents_with_retries

src/test/spec/json/transactions/README.md

@@ -19,63 +19,82 @@ These tests use a cursor's address field to track which server an operation was
 driver, use command monitoring instead.
 
 1. Test that starting a new transaction on a pinned ClientSession unpins the session and normal server selection is
-   performed for the next operation.
-
-   ```python
-   @require_server_version(4, 1, 6)
-   @require_mongos_count_at_least(2)
-   def test_unpin_for_next_transaction(self):
-     # Increase localThresholdMS and wait until both nodes are discovered
-     # to avoid false positives.
-     client = MongoClient(mongos_hosts, localThresholdMS=1000)
-     wait_until(lambda: len(client.nodes) > 1)
-     # Create the collection.
-     client.test.test.insert_one({})
-     with client.start_session() as s:
-       # Session is pinned to Mongos.
-       with s.start_transaction():
-         client.test.test.insert_one({}, session=s)
-
-       addresses = set()
-       for _ in range(50):
-         with s.start_transaction():
-           cursor = client.test.test.find({}, session=s)
-           assert next(cursor)
-           addresses.add(cursor.address)
-
-       assert len(addresses) > 1
-   ```
+    performed for the next operation.
+
+    ```python
+    @require_server_version(4, 1, 6)
+    @require_mongos_count_at_least(2)
+    def test_unpin_for_next_transaction(self):
+      # Increase localThresholdMS and wait until both nodes are discovered
+      # to avoid false positives.
+      client = MongoClient(mongos_hosts, localThresholdMS=1000)
+      wait_until(lambda: len(client.nodes) > 1)
+      # Create the collection.
+      client.test.test.insert_one({})
+      with client.start_session() as s:
+        # Session is pinned to Mongos.
+        with s.start_transaction():
+          client.test.test.insert_one({}, session=s)
+
+        addresses = set()
+        for _ in range(50):
+          with s.start_transaction():
+            cursor = client.test.test.find({}, session=s)
+            assert next(cursor)
+            addresses.add(cursor.address)
+
+        assert len(addresses) > 1
+    ```
 
 2. Test non-transaction operations using a pinned ClientSession unpins the session and normal server selection is
-   performed.
-
-   ```python
-   @require_server_version(4, 1, 6)
-   @require_mongos_count_at_least(2)
-   def test_unpin_for_non_transaction_operation(self):
-     # Increase localThresholdMS and wait until both nodes are discovered
-     # to avoid false positives.
-     client = MongoClient(mongos_hosts, localThresholdMS=1000)
-     wait_until(lambda: len(client.nodes) > 1)
-     # Create the collection.
-     client.test.test.insert_one({})
-     with client.start_session() as s:
-       # Session is pinned to Mongos.
-       with s.start_transaction():
-         client.test.test.insert_one({}, session=s)
-
-       addresses = set()
-       for _ in range(50):
-         cursor = client.test.test.find({}, session=s)
-         assert next(cursor)
-         addresses.add(cursor.address)
-
-       assert len(addresses) > 1
-   ```
+    performed.
+
+    ```python
+    @require_server_version(4, 1, 6)
+    @require_mongos_count_at_least(2)
+    def test_unpin_for_non_transaction_operation(self):
+      # Increase localThresholdMS and wait until both nodes are discovered
+      # to avoid false positives.
+      client = MongoClient(mongos_hosts, localThresholdMS=1000)
+      wait_until(lambda: len(client.nodes) > 1)
+      # Create the collection.
+      client.test.test.insert_one({})
+      with client.start_session() as s:
+        # Session is pinned to Mongos.
+        with s.start_transaction():
+          client.test.test.insert_one({}, session=s)
+
+        addresses = set()
+        for _ in range(50):
+          cursor = client.test.test.find({}, session=s)
+          assert next(cursor)
+          addresses.add(cursor.address)
+
+        assert len(addresses) > 1
+    ```
+
+3. Test that `PoolClearedError` has `TransientTransactionError` label. Since there is no simple way to trigger
+    `PoolClearedError`, this test should be implemented in a way that suites each driver the best.
+
+## Options Inside Transaction Prose Tests.
+
+These prose tests ensure drivers handle options inside a transaction where the unified tests do not suffice. Ensure
+these tests do not run against a standalone server.
+
+### 1.0 Write concern not inherited from collection object inside transaction.
+
+- Create a MongoClient running against a configured sharded/replica set/load balanced cluster.
+- Start a new session on the client.
+- Start a transaction on the session.
+- Instantiate a collection object in the driver with a default write concern of `{ w: 0 }`.
+- Insert the document `{ n: 1 }` on the instantiated collection.
+- Commit the transaction.
+- End the session.
+- Ensure the document was inserted and no error was thrown from the transaction.
 
 ## Changelog
 
+- 2024-10-31: Add test for PoolClearedError.
 - 2024-02-15: Migrated from reStructuredText to Markdown.
-- 2024-02-07: Converted legacy transaction tests to unified format and moved the\
-  legacy test format docs to a separate
-  file.
+- 2024-02-07: Converted legacy transaction tests to unified format and moved the legacy test format docs to a separate
+    file.