bahman/scylla
1
0
Fork 0
Code Issues Pull Requests Actions 3 Packages Projects Releases Wiki Activity

test: fix flaky test_read_repair_with_trace_logging by reading tracing with CL=ALL

Browse Source 
Tracing events are written to system_traces.events with CL=ANY, so they
are only guaranteed to be present on the local node of the query
coordinator. Reading them back with the driver default (CL=LOCAL_ONE)
may route the query to a replica that has not yet received all events,
causing the assertion on 'digest mismatch, starting read repair' to fail
intermittently.

Fix execute_with_tracing() to read tracing via the ResponseFuture API
with query_cl=ConsistencyLevel.ALL, so events from all replicas are
merged before the caller inspects them.

Fixes: SCYLLADB-1633

Closes scylladb/scylladb#29566
This commit is contained in:
Botond Dénes
2026-04-20 16:35:15 +03:00
committed by Patryk Jędrzejczak
parent 878f341338
commit b49cf6247f
1 changed files with 6 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
test/pylib/util.py
View File

@@ -425,7 +425,12 @@ def execute_with_tracing(cql : Session, statement : str | Statement, log : bool
cql_execute_extra_kwargs['trace'] = True
query_result = cql.execute(statement, *cql_execute_extra_args, **cql_execute_extra_kwargs)
tracing = query_result.get_all_query_traces(max_wait_sec_per=900)
# Tracing events are written to system_traces.events with CL=ANY, so they are
# only guaranteed to be present on the local node of the query coordinator.
# Reading them back with the driver default (CL=LOCAL_ONE) may hit a replica
# that has not yet received all events, causing intermittent failures.
# Using CL=ALL ensures events from all replicas are merged.
tracing = query_result.response_future.get_all_query_traces(max_wait_per=900, query_cl=ConsistencyLevel.ALL)
ret = []
page_traces = []