tests/integration: do not split paramaterized tests into multiple groups

There is also one assumption the runner already, that leads to
misleading reports like in [1].

  [1]: https://s3.amazonaws.com/clickhouse-test-reports/44762/89c071e291980e355f2c2be42dc15047caabc9bc/integration_tests__asan__%5B4/6%5D.html

Here the test test_multiple_disks/test.py::test_jbod_overflow had been
splitted into multiple groups, however because of that one assumption
it had been runned as a whole in both groups and in one it failed but
succeeded in another.

Signed-off-by: Azat Khuzhin <a.khuzhin@semrush.com>
This commit is contained in:
Azat Khuzhin 2023-01-01 23:35:21 +01:00
parent 700943a2b0
commit 4b71b481c7

View File

@ -382,7 +382,7 @@ class ClickhouseIntegrationTestsRunner:
cmd = (
"cd {repo_path}/tests/integration && "
"timeout -s 9 1h ./runner {runner_opts} {image_cmd} ' --setup-plan' "
"| tee {out_file_full} | grep '::' | sed 's/ (fixtures used:.*//g' | sed 's/^ *//g' | sed 's/ *$//g' "
"| tee {out_file_full} | grep '::' | sed 's/ (fixtures used:.*//g' | sed 's/^ *//g' | sed 's/ *$//g' | sed 's/\[.*$//g' "
"| grep -v 'SKIPPED' | sort -u > {out_file}".format(
repo_path=repo_path,
runner_opts=self._get_runner_opts(),