This is an automated email from the ASF dual-hosted git repository.

netudima pushed a commit to branch trunk
in repository https://gitbox.apache.org/repos/asf/cassandra-dtest.git

commit 26eb477126e8d4a4be11d276ae5225cc5dcffbde
Author: Dmitry Konstantinov <[email protected]>
AuthorDate: Thu Oct 29 12:19:53 2026 +0000

    Raise the default driver request timeout in dtests to 30s
    
    A DDL's client request timeout also covers the driver's wait for schema 
agreement. On CI agents with slow fsync this can exceed the driver's default of 
10s. A DROP flushes the snapshot and the commitlog-dirty tables on every node, 
one node after another. The first CREATE KEYSPACE on a fresh multi-node cluster 
queues behind no-op schema pulls. Tests such as test_6924_dropping_cf/ks, 
test_drop_mv and test_rf_expand_gossiping_property_file_snitch_multi_dc then 
fail with OperationTimedOut [...]
    
    patch by Dmitry Konstantinov; reviewed by Francisco Guerrero for 
CASSANDRA-21820
---
 dtest.py | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/dtest.py b/dtest.py
index 39ddba97..485a9e65 100644
--- a/dtest.py
+++ b/dtest.py
@@ -159,15 +159,22 @@ class Runner(threading.Thread):
             raise self.__error
 
 
-def make_execution_profile(retry_policy=FlakyRetryPolicy(), 
consistency_level=ConsistencyLevel.ONE, **kwargs):
+# Use a 30s request timeout instead of the driver default of 10s.
+# For a schema change, the driver waits for all nodes to agree on the new 
schema before
+# the request completes, and this wait counts against the request timeout.
+# On CI agents with slow disks, a DROP or the first CREATE KEYSPACE in a new 
cluster can take longer than 10s.
+def make_execution_profile(retry_policy=FlakyRetryPolicy(), 
consistency_level=ConsistencyLevel.ONE,
+                           request_timeout=30, **kwargs):
     if 'load_balancing_policy' in kwargs:
         return ExecutionProfile(retry_policy=retry_policy,
                                 consistency_level=consistency_level,
+                                request_timeout=request_timeout,
                                 **kwargs)
     else:
         return ExecutionProfile(retry_policy=retry_policy,
                                 consistency_level=consistency_level,
                                 load_balancing_policy=RoundRobinPolicy(),
+                                request_timeout=request_timeout,
                                 **kwargs)
 
 


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to