[Stateless tests (debug, s3 storage) [3/3] — Tests are not finished, fail: 1, passed: 1362, skipped: 24]https://s3.amazonaws.com/clickhouse-test-reports/41178/88d55330428eeff52dd6ec21de4f867cd45c2ba7/stateless_tests__debug__s3_storage__[3/3].html)

• WriteBufferFromS3 is not finalized in destructor. It's a bug #38158

Stateless tests (debug, s3 storage) [1/3] — fail: 1, passed: 1465, skipped: 32

• 00632_get_sample_block_cache timeout

I'm not sure that making the query single-threaded is a better solution, but I also have nothing against it

For this particular test I think it is fine, since it is about memory consumption, more details can be found here #11022
Actually while rewriting the test in #11119 it looses it's ability to reproduce the problem, will fix it.

Test failures looks unrelated:

Integration tests (release) [2/2] — fail: 1, passed: 1133, flaky: 0

• test_cluster_copier/test.py::test_copy_with_recovering[False] - Flaky test: test_cluster_copier #30399

Stateful tests (debug) — fail: 1, passed: 127, skipped: 2

• 00168_parallel_processing_on_replicas_part_1 - timeout, x2, 1200 sec instead of 600

Stateless tests (debug, s3 storage) [2/3] — fail: 3, passed: 1437, skipped: 37

• 01516_drop_table_stress_long - timeout
• 01195_formats_diagnostic_info - timeout
• 01085_max_distributed_connections - timeout