Philip Thompson created CASSANDRA-10858:
-------------------------------------------
Summary: test_copy_to_with_more_failures_than_max_attempts is
failing
Key: CASSANDRA-10858
URL: https://issues.apache.org/jira/browse/CASSANDRA-10858
Project: Cassandra
Issue Type: Sub-task
Components: Testing, Tools
Reporter: Philip Thompson
Fix For: 2.1.x, 2.2.x, 3.0.x, 3.x
{{cqlsh_tests.cqlsh_copy_tests.CqlshCopyTest.test_copy_to_with_more_failures_than_max_attempts}}
which was introduced for CASSANDRA-9306 is failing when run on clusters
without vnodes. To reproduce, simply run the test with the environment variable
DISABLE_VNODES=true.
Here is the entire debug output:
{code}
dtest: DEBUG: cluster ccm directory:
/var/folders/v3/z4wf_34n1q506_xjdy49gb780000gn/T/dtest-iJQECY
dtest: DEBUG: removing ccm cluster test at:
/var/folders/v3/z4wf_34n1q506_xjdy49gb780000gn/T/dtest-iJQECY
dtest: DEBUG: clearing ssl stores from
[/var/folders/v3/z4wf_34n1q506_xjdy49gb780000gn/T/dtest-iJQECY] directory
dtest: DEBUG: cluster ccm directory:
/var/folders/v3/z4wf_34n1q506_xjdy49gb780000gn/T/dtest-CnOgA8
dtest: DEBUG: Running stress
dtest: DEBUG: Exporting to csv file:
/var/folders/v3/z4wf_34n1q506_xjdy49gb780000gn/T/tmpZYqB01 with
{"failing_range": {"start": 0, "end": 5000000000000000000, "num_failures": 5}}
and 3 max attempts
dtest: DEBUG:
Starting copy of keyspace1.standard1 with columns ['key', 'C0', 'C1', 'C2',
'C3', 'C4'].
Processed 10000 rows; Written: 10503.508303 rows/s
Processed 20000 rows; Written: 11860.954046 rows/s
Processed 30000 rows; Written: 13068.388704 rows/s
Processed 40000 rows; Written: 16941.628006 rows/s
Processed 50000 rows; Written: 17609.109488 rows/s
Processed 60000 rows; Written: 19475.156238 rows/s
Processed 70000 rows; Written: 19976.978154 rows/s
Processed 80000 rows; Written: 19992.329090 rows/s
Processed 90000 rows; Written: 20623.291907 rows/s
Processed 100000 rows; Written: 21644.815363 rows/s
Processed 100000 rows; Written: 10822.407682 rows/s
100000 rows exported in 5.816 seconds.
{code}
I assume this is related to the failure injection code in cqlsh not handling
fewer token ranges.
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
